{"id": "scene0006_00_0", "scene_id": "scene0006_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": "validation"}
{"id": "scene0006_00_1", "scene_id": "scene0006_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 bag appear?", "choices": null, "answer": "Yes", "target": "bag", "split": "validation"}
{"id": "scene0006_00_2", "scene_id": "scene0006_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 bag in this scene?\nA. lamp\nB. backpack\nC. nightstand\nD. shoes", "choices": ["A. lamp", "B. backpack", "C. nightstand", "D. shoes"], "answer": "D. shoes", "target": "bag", "split": "validation"}
{"id": "scene0006_00_3", "scene_id": "scene0006_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 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 6 cm away.", "target": "bag", "split": "validation"}
{"id": "scene0006_00_4", "scene_id": "scene0006_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 bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"suitcase\", \"references\": [{\"object\": \"shoes\", \"relation\": \"side_above\", \"distance_cm\": 6.3}]}", "target": "bag", "split": "validation"}
{"id": "scene0006_00_5", "scene_id": "scene0006_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": "validation"}
{"id": "scene0006_00_6", "scene_id": "scene0006_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 purse in this scene?\nA. lamp\nB. bag\nC. backpack\nD. suitcase", "choices": ["A. lamp", "B. bag", "C. backpack", "D. suitcase"], "answer": "D. suitcase", "target": "purse", "split": "validation"}
{"id": "scene0006_00_7", "scene_id": "scene0006_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 purse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The purse is near the suitcase, about 23 cm away.", "target": "purse", "split": "validation"}
{"id": "scene0006_00_8", "scene_id": "scene0006_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 purse in structured JSON format.", "choices": null, "answer": "{\"target\": \"purse\", \"support_surface\": \"backpack\", \"references\": [{\"object\": \"suitcase\", \"relation\": \"near\", \"distance_cm\": 23.4}]}", "target": "purse", "split": "validation"}
{"id": "scene0006_00_9", "scene_id": "scene0006_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 toaster oven appear?", "choices": null, "answer": "No", "target": "toaster oven", "split": "validation"}
{"id": "scene0006_00_10", "scene_id": "scene0006_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. backpack\nB. lamp\nC. pillow\nD. bag", "choices": ["A. backpack", "B. lamp", "C. pillow", "D. bag"], "answer": "B. lamp", "target": "telephone", "split": "validation"}
{"id": "scene0006_00_11", "scene_id": "scene0006_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 nightstand. It is also to the side of and below the lamp, about 5 cm away. It is further to the side of and below the pillow, about 20 cm away.", "target": "telephone", "split": "validation"}
{"id": "scene0006_00_12", "scene_id": "scene0006_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\": \"nightstand\", \"references\": [{\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 5.0}, {\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 19.7}]}", "target": "telephone", "split": "validation"}
{"id": "scene0024_00_0", "scene_id": "scene0024_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": "validation"}
{"id": "scene0024_00_1", "scene_id": "scene0024_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 bag appear?", "choices": null, "answer": "Yes", "target": "bag", "split": "validation"}
{"id": "scene0024_00_2", "scene_id": "scene0024_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 bag in this scene?\nA. book\nB. purse\nC. backpack\nD. chair", "choices": ["A. book", "B. purse", "C. backpack", "D. chair"], "answer": "A. book", "target": "bag", "split": "validation"}
{"id": "scene0024_00_3", "scene_id": "scene0024_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 bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The bag is on the table. It is also next to the book, about 15 cm away. It is further to the side of and above the purse, about 59 cm away.", "target": "bag", "split": "validation"}
{"id": "scene0024_00_4", "scene_id": "scene0024_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 bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"bag\", \"support_surface\": \"table\", \"references\": [{\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 14.5}, {\"object\": \"purse\", \"relation\": \"side_above\", \"distance_cm\": 59.1}]}", "target": "bag", "split": "validation"}
{"id": "scene0024_00_5", "scene_id": "scene0024_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 ball appear?", "choices": null, "answer": "Yes", "target": "ball", "split": "validation"}
{"id": "scene0024_00_6", "scene_id": "scene0024_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 ball in this scene?\nA. table\nB. plant\nC. lamp\nD. book", "choices": ["A. table", "B. plant", "C. lamp", "D. book"], "answer": "A. table", "target": "ball", "split": "validation"}
{"id": "scene0024_00_7", "scene_id": "scene0024_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 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 table, about 9 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 below the plant, about 36 cm away.", "target": "ball", "split": "validation"}
{"id": "scene0024_00_8", "scene_id": "scene0024_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 ball in structured JSON format.", "choices": null, "answer": "{\"target\": \"ball\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 8.7}, {\"object\": \"book\", \"relation\": \"side_below\", \"distance_cm\": 31.0}]}", "target": "ball", "split": "validation"}
{"id": "scene0024_00_9", "scene_id": "scene0024_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 coffee kettle appear?", "choices": null, "answer": "No", "target": "coffee kettle", "split": "validation"}
{"id": "scene0024_00_10", "scene_id": "scene0024_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 purse in this scene?\nA. book\nB. backpack\nC. bag\nD. chair", "choices": ["A. book", "B. backpack", "C. bag", "D. chair"], "answer": "A. book", "target": "purse", "split": "validation"}
{"id": "scene0024_00_11", "scene_id": "scene0024_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 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 book, about 25 cm away. It is also to the side of and below the bag, about 59 cm away.", "target": "purse", "split": "validation"}
{"id": "scene0024_00_12", "scene_id": "scene0024_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 purse in structured JSON format.", "choices": null, "answer": "{\"target\": \"purse\", \"support_surface\": \"chair\", \"references\": [{\"object\": \"book\", \"relation\": \"side_below\", \"distance_cm\": 24.6}, {\"object\": \"bag\", \"relation\": \"side_below\", \"distance_cm\": 59.1}]}", "target": "purse", "split": "validation"}
{"id": "scene0025_00_0", "scene_id": "scene0025_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": "validation"}
{"id": "scene0025_00_1", "scene_id": "scene0025_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\": \"cabinet\", \"references\": []}", "target": "book", "split": "validation"}
{"id": "scene0025_00_2", "scene_id": "scene0025_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 shoe appear?", "choices": null, "answer": "No", "target": "shoe", "split": "validation"}
{"id": "scene0025_00_3", "scene_id": "scene0025_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 headphones in this scene?\nA. keyboard\nB. office chair\nC. monitor\nD. desk", "choices": ["A. keyboard", "B. office chair", "C. monitor", "D. desk"], "answer": "A. keyboard", "target": "headphones", "split": "validation"}
{"id": "scene0025_00_4", "scene_id": "scene0025_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 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 keyboard, about 3 cm away. It is further to the side of and below the monitor, about 5 cm away.", "target": "headphones", "split": "validation"}
{"id": "scene0025_00_5", "scene_id": "scene0025_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 headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 2.9}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.5}]}", "target": "headphones", "split": "validation"}
{"id": "scene0025_00_6", "scene_id": "scene0025_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 water pitcher appear?", "choices": null, "answer": "No", "target": "water pitcher", "split": "validation"}
{"id": "scene0025_00_7", "scene_id": "scene0025_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 paper towel roll in this scene?\nA. keyboard\nB. monitor\nC. office chair\nD. desk", "choices": ["A. keyboard", "B. monitor", "C. office chair", "D. desk"], "answer": "D. desk", "target": "paper towel roll", "split": "validation"}
{"id": "scene0025_00_8", "scene_id": "scene0025_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 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 desk, about 13 cm away. It is also next to the keyboard, about 15 cm away. It is further to the side of and above the office chair, about 49 cm away.", "target": "paper towel roll", "split": "validation"}
{"id": "scene0025_00_9", "scene_id": "scene0025_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 paper towel roll in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper towel roll\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 12.5}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 14.8}]}", "target": "paper towel roll", "split": "validation"}
{"id": "scene0025_00_10", "scene_id": "scene0025_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 telephone appear?", "choices": null, "answer": "Yes", "target": "telephone", "split": "validation"}
{"id": "scene0025_00_11", "scene_id": "scene0025_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 telephone in this scene?\nA. office chair\nB. monitor\nC. box\nD. desk", "choices": ["A. office chair", "B. monitor", "C. box", "D. desk"], "answer": "C. box", "target": "telephone", "split": "validation"}
{"id": "scene0025_00_12", "scene_id": "scene0025_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 telephone in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The telephone is near the monitor, about 22 cm away. It is also to the side of and above the desk, about 28 cm away. It is further to the side of and above the office chair, about 54 cm away.", "target": "telephone", "split": "validation"}
{"id": "scene0025_00_13", "scene_id": "scene0025_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 telephone in structured JSON format.", "choices": null, "answer": "{\"target\": \"telephone\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"near\", \"distance_cm\": 21.6}, {\"object\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 28.0}]}", "target": "telephone", "split": "validation"}
{"id": "scene0034_00_0", "scene_id": "scene0034_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": "validation"}
{"id": "scene0034_00_1", "scene_id": "scene0034_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. mirror\nB. sink\nC. scale\nD. bathroom vanity", "choices": ["A. mirror", "B. sink", "C. scale", "D. bathroom vanity"], "answer": "A. mirror", "target": "cup", "split": "validation"}
{"id": "scene0034_00_2", "scene_id": "scene0034_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 mirror, about 34 cm away.", "target": "cup", "split": "validation"}
{"id": "scene0034_00_3", "scene_id": "scene0034_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\": \"bathroom vanity\", \"references\": [{\"object\": \"mirror\", \"relation\": \"side_below\", \"distance_cm\": 33.9}]}", "target": "cup", "split": "validation"}
{"id": "scene0034_00_4", "scene_id": "scene0034_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": "validation"}
{"id": "scene0034_00_5", "scene_id": "scene0034_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 scale in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The scale is below the towel, about 17 cm away. It is also to the side of and below the shelf, about 17 cm away.", "target": "scale", "split": "validation"}
{"id": "scene0034_00_6", "scene_id": "scene0034_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 scale in structured JSON format.", "choices": null, "answer": "{\"target\": \"scale\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"towel\", \"relation\": \"below\", \"distance_cm\": 17.0}, {\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 17.0}]}", "target": "scale", "split": "validation"}
{"id": "scene0034_00_7", "scene_id": "scene0034_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 paper bag appear?", "choices": null, "answer": "No", "target": "paper bag", "split": "validation"}
{"id": "scene0034_00_8", "scene_id": "scene0034_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 toilet paper in this scene?\nA. toilet\nB. bathroom vanity\nC. trash bin\nD. cup", "choices": ["A. toilet", "B. bathroom vanity", "C. trash bin", "D. cup"], "answer": "A. toilet", "target": "toilet paper", "split": "validation"}
{"id": "scene0034_00_9", "scene_id": "scene0034_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 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 43 cm away. It is also to the side of and above the trash bin, about 72 cm away.", "target": "toilet paper", "split": "validation"}
{"id": "scene0034_00_10", "scene_id": "scene0034_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 toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"trash bin\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 42.7}, {\"object\": \"trash bin\", \"relation\": \"side_above\", \"distance_cm\": 71.9}]}", "target": "toilet paper", "split": "validation"}
{"id": "scene0040_00_0", "scene_id": "scene0040_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": "validation"}
{"id": "scene0040_00_1", "scene_id": "scene0040_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 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 box, about 11 cm away. It is further to the side of and above the file cabinet, about 12 cm away.", "target": "bottle", "split": "validation"}
{"id": "scene0040_00_2", "scene_id": "scene0040_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 bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 11.2}, {\"object\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 12.1}]}", "target": "bottle", "split": "validation"}
{"id": "scene0040_00_3", "scene_id": "scene0040_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 paper cutter appear?", "choices": null, "answer": "No", "target": "paper cutter", "split": "validation"}
{"id": "scene0040_00_4", "scene_id": "scene0040_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 clock in this scene?\nA. bottle\nB. monitor\nC. box\nD. books", "choices": ["A. bottle", "B. monitor", "C. box", "D. books"], "answer": "B. monitor", "target": "clock", "split": "validation"}
{"id": "scene0040_00_5", "scene_id": "scene0040_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 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 monitor, about 53 cm away. It is also to the side of and above the books, about 63 cm away.", "target": "clock", "split": "validation"}
{"id": "scene0040_00_6", "scene_id": "scene0040_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 clock in structured JSON format.", "choices": null, "answer": "{\"target\": \"clock\", \"support_surface\": \"books\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_above\", \"distance_cm\": 52.5}, {\"object\": \"books\", \"relation\": \"side_above\", \"distance_cm\": 63.5}]}", "target": "clock", "split": "validation"}
{"id": "scene0040_00_7", "scene_id": "scene0040_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 soap dish appear?", "choices": null, "answer": "No", "target": "soap dish", "split": "validation"}
{"id": "scene0040_00_8", "scene_id": "scene0040_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 headphones in this scene?\nA. file cabinet\nB. teddy bear\nC. keyboard\nD. cup", "choices": ["A. file cabinet", "B. teddy bear", "C. keyboard", "D. cup"], "answer": "C. keyboard", "target": "headphones", "split": "validation"}
{"id": "scene0040_00_9", "scene_id": "scene0040_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 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 keyboard, about 12 cm away. It is further near the cup, about 17 cm away.", "target": "headphones", "split": "validation"}
{"id": "scene0040_00_10", "scene_id": "scene0040_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 headphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphones\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"side_above\", \"distance_cm\": 11.9}, {\"object\": \"cup\", \"relation\": \"near\", \"distance_cm\": 16.9}]}", "target": "headphones", "split": "validation"}
{"id": "scene0040_00_11", "scene_id": "scene0040_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 mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "validation"}
{"id": "scene0040_00_12", "scene_id": "scene0040_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 mouse in this scene?\nA. file cabinet\nB. box\nC. bottle\nD. chair", "choices": ["A. file cabinet", "B. box", "C. bottle", "D. chair"], "answer": "B. box", "target": "mouse", "split": "validation"}
{"id": "scene0040_00_13", "scene_id": "scene0040_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 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 box, about 22 cm away. It is further near the bottle, about 29 cm away.", "target": "mouse", "split": "validation"}
{"id": "scene0040_00_14", "scene_id": "scene0040_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 mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 22.4}, {\"object\": \"bottle\", \"relation\": \"near\", \"distance_cm\": 28.5}]}", "target": "mouse", "split": "validation"}
{"id": "scene0040_00_15", "scene_id": "scene0040_00", "source_dataset": "scannet200", "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 cd case appear?", "choices": null, "answer": "No", "target": "cd case", "split": "validation"}
{"id": "scene0040_00_16", "scene_id": "scene0040_00", "source_dataset": "scannet200", "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 paper bag in this scene?\nA. box\nB. light switch\nC. books\nD. bottle", "choices": ["A. box", "B. light switch", "C. books", "D. bottle"], "answer": "B. light switch", "target": "paper bag", "split": "validation"}
{"id": "scene0040_00_17", "scene_id": "scene0040_00", "source_dataset": "scannet200", "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 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 light switch, about 69 cm away.", "target": "paper bag", "split": "validation"}
{"id": "scene0040_00_18", "scene_id": "scene0040_00", "source_dataset": "scannet200", "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 paper bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper bag\", \"support_surface\": \"light switch\", \"references\": [{\"object\": \"light switch\", \"relation\": \"side_below\", \"distance_cm\": 68.6}]}", "target": "paper bag", "split": "validation"}
{"id": "scene0040_00_19", "scene_id": "scene0040_00", "source_dataset": "scannet200", "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 ball appear?", "choices": null, "answer": "No", "target": "ball", "split": "validation"}
{"id": "scene0040_00_20", "scene_id": "scene0040_00", "source_dataset": "scannet200", "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 stuffed animal in this scene?\nA. file cabinet\nB. monitor\nC. box\nD. books", "choices": ["A. file cabinet", "B. monitor", "C. box", "D. books"], "answer": "D. books", "target": "stuffed animal", "split": "validation"}
{"id": "scene0040_00_21", "scene_id": "scene0040_00", "source_dataset": "scannet200", "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 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 books, about 9 cm away. It is further to the side of and above the file cabinet, about 37 cm away.", "target": "stuffed animal", "split": "validation"}
{"id": "scene0040_00_22", "scene_id": "scene0040_00", "source_dataset": "scannet200", "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 stuffed animal in structured JSON format.", "choices": null, "answer": "{\"target\": \"stuffed animal\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"books\", \"relation\": \"next_to\", \"distance_cm\": 8.8}, {\"object\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 36.6}]}", "target": "stuffed animal", "split": "validation"}
{"id": "scene0042_00_0", "scene_id": "scene0042_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": "validation"}
{"id": "scene0042_00_1", "scene_id": "scene0042_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. chair\nB. mirror\nC. sink\nD. monitor", "choices": ["A. chair", "B. mirror", "C. sink", "D. monitor"], "answer": "C. sink", "target": "soap dispenser", "split": "validation"}
{"id": "scene0042_00_2", "scene_id": "scene0042_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 29 cm away. It is also near the mirror, about 48 cm away.", "target": "soap dispenser", "split": "validation"}
{"id": "scene0042_00_3", "scene_id": "scene0042_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\": 28.6}, {\"object\": \"mirror\", \"relation\": \"near\", \"distance_cm\": 47.7}]}", "target": "soap dispenser", "split": "validation"}
{"id": "scene0047_00_0", "scene_id": "scene0047_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": "validation"}
{"id": "scene0047_00_1", "scene_id": "scene0047_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. microwave\nB. kitchen cabinets\nC. sink\nD. paper towel dispenser", "choices": ["A. microwave", "B. kitchen cabinets", "C. sink", "D. paper towel dispenser"], "answer": "D. paper towel dispenser", "target": "soap dispenser", "split": "validation"}
{"id": "scene0047_00_2", "scene_id": "scene0047_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 paper towel dispenser, about 6 cm away. It is also to the side of and above the sink, about 19 cm away. It is further to the side of and below the kitchen cabinets, about 26 cm away.", "target": "soap dispenser", "split": "validation"}
{"id": "scene0047_00_3", "scene_id": "scene0047_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\": \"paper towel dispenser\", \"relation\": \"next_to\", \"distance_cm\": 5.5}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 19.4}]}", "target": "soap dispenser", "split": "validation"}
{"id": "scene0048_00_0", "scene_id": "scene0048_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": "validation"}
{"id": "scene0048_00_1", "scene_id": "scene0048_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. table\nD. desk", "choices": ["A. chair", "B. monitor", "C. table", "D. desk"], "answer": "A. chair", "target": "backpack", "split": "validation"}
{"id": "scene0048_00_2", "scene_id": "scene0048_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 54 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0048_00_3", "scene_id": "scene0048_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\": 53.7}]}", "target": "backpack", "split": "validation"}
{"id": "scene0070_00_0", "scene_id": "scene0070_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": "validation"}
{"id": "scene0070_00_1", "scene_id": "scene0070_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. chair\nB. picture\nC. file cabinet\nD. ledge", "choices": ["A. chair", "B. picture", "C. file cabinet", "D. ledge"], "answer": "B. picture", "target": "clock", "split": "validation"}
{"id": "scene0070_00_2", "scene_id": "scene0070_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 picture, about 13 cm away. It is also to the side of and above the ledge, about 67 cm away.", "target": "clock", "split": "validation"}
{"id": "scene0070_00_3", "scene_id": "scene0070_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\": \"picture\", \"references\": [{\"object\": \"picture\", \"relation\": \"side_above\", \"distance_cm\": 13.1}, {\"object\": \"ledge\", \"relation\": \"side_above\", \"distance_cm\": 67.1}]}", "target": "clock", "split": "validation"}
{"id": "scene0070_00_4", "scene_id": "scene0070_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": "validation"}
{"id": "scene0070_00_5", "scene_id": "scene0070_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. file cabinet\nD. ledge", "choices": ["A. monitor", "B. chair", "C. file cabinet", "D. ledge"], "answer": "C. file cabinet", "target": "keyboard", "split": "validation"}
{"id": "scene0070_00_6", "scene_id": "scene0070_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 table. 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 26 cm away.", "target": "keyboard", "split": "validation"}
{"id": "scene0070_00_7", "scene_id": "scene0070_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\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 18.0}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 25.8}]}", "target": "keyboard", "split": "validation"}
{"id": "scene0083_00_0", "scene_id": "scene0083_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": "validation"}
{"id": "scene0083_00_1", "scene_id": "scene0083_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. monitor\nB. toilet paper\nC. toilet\nD. bathtub", "choices": ["A. monitor", "B. toilet paper", "C. toilet", "D. bathtub"], "answer": "D. bathtub", "target": "soap dish", "split": "validation"}
{"id": "scene0083_00_2", "scene_id": "scene0083_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 8 cm away.", "target": "soap dish", "split": "validation"}
{"id": "scene0083_00_3", "scene_id": "scene0083_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\": 8.3}]}", "target": "soap dish", "split": "validation"}
{"id": "scene0083_00_4", "scene_id": "scene0083_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": "validation"}
{"id": "scene0083_00_5", "scene_id": "scene0083_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. monitor\nB. toilet\nC. bathtub\nD. soap dish", "choices": ["A. monitor", "B. toilet", "C. bathtub", "D. soap dish"], "answer": "B. toilet", "target": "toilet paper", "split": "validation"}
{"id": "scene0083_00_6", "scene_id": "scene0083_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 22 cm away.", "target": "toilet paper", "split": "validation"}
{"id": "scene0083_00_7", "scene_id": "scene0083_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\": 21.9}]}", "target": "toilet paper", "split": "validation"}
{"id": "scene0084_00_0", "scene_id": "scene0084_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": "validation"}
{"id": "scene0084_00_1", "scene_id": "scene0084_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. mirror\nB. bathroom counter\nC. sink\nD. sign", "choices": ["A. mirror", "B. bathroom counter", "C. sink", "D. sign"], "answer": "B. bathroom counter", "target": "soap dispenser", "split": "validation"}
{"id": "scene0084_00_2", "scene_id": "scene0084_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 bathroom counter, about 3 cm away. It is also to the side of and above the sink, about 19 cm away. It is further near the mirror, about 42 cm away.", "target": "soap dispenser", "split": "validation"}
{"id": "scene0084_00_3", "scene_id": "scene0084_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 counter\", \"references\": [{\"object\": \"bathroom counter\", \"relation\": \"side_above\", \"distance_cm\": 3.2}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 19.2}]}", "target": "soap dispenser", "split": "validation"}
{"id": "scene0084_00_4", "scene_id": "scene0084_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": "validation"}
{"id": "scene0084_00_5", "scene_id": "scene0084_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 counter\nB. jacket\nC. toilet\nD. mirror", "choices": ["A. bathroom counter", "B. jacket", "C. toilet", "D. mirror"], "answer": "C. toilet", "target": "toilet paper", "split": "validation"}
{"id": "scene0084_00_6", "scene_id": "scene0084_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.", "target": "toilet paper", "split": "validation"}
{"id": "scene0084_00_7", "scene_id": "scene0084_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 paper holder\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 27.9}]}", "target": "toilet paper", "split": "validation"}
{"id": "scene0086_00_0", "scene_id": "scene0086_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": "validation"}
{"id": "scene0086_00_1", "scene_id": "scene0086_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. trash can\nB. paper towel dispenser\nC. plant\nD. mirror", "choices": ["A. trash can", "B. paper towel dispenser", "C. plant", "D. mirror"], "answer": "A. trash can", "target": "bottle", "split": "validation"}
{"id": "scene0086_00_2", "scene_id": "scene0086_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 trash can, about 24 cm away. It is also to the side of and above the paper towel dispenser, about 68 cm away.", "target": "bottle", "split": "validation"}
{"id": "scene0086_00_3", "scene_id": "scene0086_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\": \"paper towel dispenser\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 24.1}, {\"object\": \"paper towel dispenser\", \"relation\": \"side_above\", \"distance_cm\": 68.0}]}", "target": "bottle", "split": "validation"}
{"id": "scene0086_00_4", "scene_id": "scene0086_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": "validation"}
{"id": "scene0086_00_5", "scene_id": "scene0086_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. bottle\nB. plant\nC. sink\nD. paper towel dispenser", "choices": ["A. bottle", "B. plant", "C. sink", "D. paper towel dispenser"], "answer": "C. sink", "target": "soap dispenser", "split": "validation"}
{"id": "scene0086_00_6", "scene_id": "scene0086_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 sink, about 7 cm away. It is also to the side of and above the paper towel dispenser, about 44 cm away. It is further to the side of and below the plant, about 69 cm away.", "target": "soap dispenser", "split": "validation"}
{"id": "scene0086_00_7", "scene_id": "scene0086_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\": \"paper towel dispenser\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 7.0}, {\"object\": \"paper towel dispenser\", \"relation\": \"side_above\", \"distance_cm\": 43.5}]}", "target": "soap dispenser", "split": "validation"}
{"id": "scene0089_00_0", "scene_id": "scene0089_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": "No", "target": "backpack", "split": "validation"}
{"id": "scene0089_00_1", "scene_id": "scene0089_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. desk\nB. monitor\nC. radiator\nD. chair", "choices": ["A. desk", "B. monitor", "C. radiator", "D. chair"], "answer": "D. chair", "target": "mouse", "split": "validation"}
{"id": "scene0089_00_2", "scene_id": "scene0089_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 to the side of and above the chair, about 7 cm away.", "target": "mouse", "split": "validation"}
{"id": "scene0089_00_3", "scene_id": "scene0089_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\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 7.5}]}", "target": "mouse", "split": "validation"}
{"id": "scene0095_00_0", "scene_id": "scene0095_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": "Yes", "target": "keyboard", "split": "validation"}
{"id": "scene0095_00_1", "scene_id": "scene0095_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. desk\nC. mouse\nD. monitor", "choices": ["A. chair", "B. desk", "C. mouse", "D. monitor"], "answer": "D. monitor", "target": "keyboard", "split": "validation"}
{"id": "scene0095_00_2", "scene_id": "scene0095_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 9 cm away. It is also next to the mouse, about 12 cm away. It is further to the side of and above the chair, about 21 cm away.", "target": "keyboard", "split": "validation"}
{"id": "scene0095_00_3", "scene_id": "scene0095_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\": 9.0}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 12.2}]}", "target": "keyboard", "split": "validation"}
{"id": "scene0095_00_4", "scene_id": "scene0095_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": "validation"}
{"id": "scene0095_00_5", "scene_id": "scene0095_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. monitor\nB. keyboard\nC. desk\nD. chair", "choices": ["A. monitor", "B. keyboard", "C. desk", "D. chair"], "answer": "D. chair", "target": "mouse", "split": "validation"}
{"id": "scene0095_00_6", "scene_id": "scene0095_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 to the side of and above the chair, about 11 cm away. It is also next to the keyboard, about 12 cm away. It is further to the side of and below the monitor, about 26 cm away.", "target": "mouse", "split": "validation"}
{"id": "scene0095_00_7", "scene_id": "scene0095_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\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 11.1}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 12.2}]}", "target": "mouse", "split": "validation"}
{"id": "scene0097_00_0", "scene_id": "scene0097_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": "validation"}
{"id": "scene0097_00_1", "scene_id": "scene0097_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 maker\nB. cabinet\nC. coffee kettle\nD. counter", "choices": ["A. coffee maker", "B. cabinet", "C. coffee kettle", "D. counter"], "answer": "B. cabinet", "target": "backpack", "split": "validation"}
{"id": "scene0097_00_2", "scene_id": "scene0097_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 34 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0097_00_3", "scene_id": "scene0097_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\": 33.6}]}", "target": "backpack", "split": "validation"}
{"id": "scene0097_00_4", "scene_id": "scene0097_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 coffee kettle appear?", "choices": null, "answer": "Yes", "target": "coffee kettle", "split": "validation"}
{"id": "scene0097_00_5", "scene_id": "scene0097_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 coffee kettle in this scene?\nA. toaster\nB. sink\nC. kitchen cabinets\nD. coffee maker", "choices": ["A. toaster", "B. sink", "C. kitchen cabinets", "D. coffee maker"], "answer": "A. toaster", "target": "coffee kettle", "split": "validation"}
{"id": "scene0097_00_6", "scene_id": "scene0097_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 coffee kettle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The coffee kettle is on the counter. It is also under the toaster, about 3 cm away. It is further to the side of and above the kitchen cabinets, about 13 cm away.", "target": "coffee kettle", "split": "validation"}
{"id": "scene0097_00_7", "scene_id": "scene0097_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 coffee kettle in structured JSON format.", "choices": null, "answer": "{\"target\": \"coffee kettle\", \"support_surface\": \"counter\", \"references\": [{\"object\": \"toaster\", \"relation\": \"under\", \"distance_cm\": 3.4}, {\"object\": \"kitchen cabinets\", \"relation\": \"side_above\", \"distance_cm\": 13.1}]}", "target": "coffee kettle", "split": "validation"}
{"id": "scene0097_00_8", "scene_id": "scene0097_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": "validation"}
{"id": "scene0097_00_9", "scene_id": "scene0097_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 in this scene?\nA. sink\nB. coffee maker\nC. coffee kettle\nD. soap", "choices": ["A. sink", "B. coffee maker", "C. coffee kettle", "D. soap"], "answer": "C. coffee kettle", "target": "toaster", "split": "validation"}
{"id": "scene0097_00_10", "scene_id": "scene0097_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 in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toaster is on the counter. It is also next to the coffee kettle, about 3 cm away. It is further to the side of and below the soap, about 6 cm away.", "target": "toaster", "split": "validation"}
{"id": "scene0097_00_11", "scene_id": "scene0097_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 in structured JSON format.", "choices": null, "answer": "{\"target\": \"toaster\", \"support_surface\": \"counter\", \"references\": [{\"object\": \"coffee kettle\", \"relation\": \"next_to\", \"distance_cm\": 3.4}, {\"object\": \"soap\", \"relation\": \"side_below\", \"distance_cm\": 5.9}]}", "target": "toaster", "split": "validation"}
{"id": "scene0100_00_0", "scene_id": "scene0100_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": "validation"}
{"id": "scene0100_00_1", "scene_id": "scene0100_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. shower control valve\nB. bathtub\nC. toilet\nD. towel", "choices": ["A. shower control valve", "B. bathtub", "C. toilet", "D. towel"], "answer": "C. toilet", "target": "toilet paper", "split": "validation"}
{"id": "scene0100_00_2", "scene_id": "scene0100_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 15 cm away. It is also to the side of and above the bathtub, about 27 cm away. It is further to the side of and below the shower control valve, about 60 cm away.", "target": "toilet paper", "split": "validation"}
{"id": "scene0100_00_3", "scene_id": "scene0100_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\": 15.5}, {\"object\": \"bathtub\", \"relation\": \"side_above\", \"distance_cm\": 26.9}]}", "target": "toilet paper", "split": "validation"}
{"id": "scene0101_00_0", "scene_id": "scene0101_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": "validation"}
{"id": "scene0101_00_1", "scene_id": "scene0101_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. backpack\nB. laptop\nC. bed\nD. chair", "choices": ["A. backpack", "B. laptop", "C. bed", "D. chair"], "answer": "C. bed", "target": "cup", "split": "validation"}
{"id": "scene0101_00_2", "scene_id": "scene0101_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 above the bed, about 48 cm away. It is further to the side of and above the chair, about 62 cm away.", "target": "cup", "split": "validation"}
{"id": "scene0101_00_3", "scene_id": "scene0101_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\": \"bed\", \"relation\": \"side_above\", \"distance_cm\": 47.9}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 61.7}]}", "target": "cup", "split": "validation"}
{"id": "scene0101_00_4", "scene_id": "scene0101_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": "validation"}
{"id": "scene0101_00_5", "scene_id": "scene0101_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 dumbbell in this scene?\nA. bed\nB. towel\nC. backpack\nD. bathroom vanity", "choices": ["A. bed", "B. towel", "C. backpack", "D. bathroom vanity"], "answer": "B. towel", "target": "dumbbell", "split": "validation"}
{"id": "scene0101_00_6", "scene_id": "scene0101_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 dumbbell in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The dumbbell is below the towel, about 41 cm away.", "target": "dumbbell", "split": "validation"}
{"id": "scene0101_00_7", "scene_id": "scene0101_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 dumbbell in structured JSON format.", "choices": null, "answer": "{\"target\": \"dumbbell\", \"support_surface\": \"stove\", \"references\": [{\"object\": \"towel\", \"relation\": \"below\", \"distance_cm\": 41.2}]}", "target": "dumbbell", "split": "validation"}
{"id": "scene0101_00_8", "scene_id": "scene0101_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": "validation"}
{"id": "scene0101_00_9", "scene_id": "scene0101_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 soap dispenser appear?", "choices": null, "answer": "No", "target": "soap dispenser", "split": "validation"}
{"id": "scene0101_00_10", "scene_id": "scene0101_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. cup\nB. trash can\nC. shelf\nD. backpack", "choices": ["A. cup", "B. trash can", "C. shelf", "D. backpack"], "answer": "C. shelf", "target": "laptop", "split": "validation"}
{"id": "scene0101_00_11", "scene_id": "scene0101_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 desk. It is also to the side of and below the shelf, about 24 cm away. It is further to the side of and above the trash can, about 65 cm away.", "target": "laptop", "split": "validation"}
{"id": "scene0101_00_12", "scene_id": "scene0101_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\": \"desk\", \"references\": [{\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 24.2}, {\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 65.4}]}", "target": "laptop", "split": "validation"}
{"id": "scene0101_00_13", "scene_id": "scene0101_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 stuffed animal appear?", "choices": null, "answer": "No", "target": "stuffed animal", "split": "validation"}
{"id": "scene0101_00_14", "scene_id": "scene0101_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 plunger in this scene?\nA. bathroom vanity\nB. toilet paper\nC. backpack\nD. toilet", "choices": ["A. bathroom vanity", "B. toilet paper", "C. backpack", "D. toilet"], "answer": "D. toilet", "target": "plunger", "split": "validation"}
{"id": "scene0101_00_15", "scene_id": "scene0101_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 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 74 cm away.", "target": "plunger", "split": "validation"}
{"id": "scene0101_00_16", "scene_id": "scene0101_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 plunger in structured JSON format.", "choices": null, "answer": "{\"target\": \"plunger\", \"support_surface\": \"toilet paper\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_below\", \"distance_cm\": 20.4}, {\"object\": \"toilet paper\", \"relation\": \"side_below\", \"distance_cm\": 74.1}]}", "target": "plunger", "split": "validation"}
{"id": "scene0101_00_17", "scene_id": "scene0101_00", "source_dataset": "scannet200", "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 toilet paper appear?", "choices": null, "answer": "Yes", "target": "toilet paper", "split": "validation"}
{"id": "scene0101_00_18", "scene_id": "scene0101_00", "source_dataset": "scannet200", "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 toilet paper in this scene?\nA. backpack\nB. toilet\nC. sink\nD. plunger", "choices": ["A. backpack", "B. toilet", "C. sink", "D. plunger"], "answer": "B. toilet", "target": "toilet paper", "split": "validation"}
{"id": "scene0101_00_19", "scene_id": "scene0101_00", "source_dataset": "scannet200", "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 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 31 cm away. It is further to the side of and above the plunger, about 74 cm away.", "target": "toilet paper", "split": "validation"}
{"id": "scene0101_00_20", "scene_id": "scene0101_00", "source_dataset": "scannet200", "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 toilet paper in structured JSON format.", "choices": null, "answer": "{\"target\": \"toilet paper\", \"support_surface\": \"bathroom vanity\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 15.6}, {\"object\": \"sink\", \"relation\": \"side_below\", \"distance_cm\": 30.6}]}", "target": "toilet paper", "split": "validation"}
{"id": "scene0109_00_0", "scene_id": "scene0109_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": "validation"}
{"id": "scene0117_00_0", "scene_id": "scene0117_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": "validation"}
{"id": "scene0117_00_1", "scene_id": "scene0117_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. tupperware\nC. cups\nD. chair", "choices": ["A. cup", "B. tupperware", "C. cups", "D. chair"], "answer": "B. tupperware", "target": "backpack", "split": "validation"}
{"id": "scene0117_00_2", "scene_id": "scene0117_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 tupperware, about 42 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0117_00_3", "scene_id": "scene0117_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\": \"tupperware\", \"relation\": \"side_below\", \"distance_cm\": 42.3}]}", "target": "backpack", "split": "validation"}
{"id": "scene0117_00_4", "scene_id": "scene0117_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 broom appear?", "choices": null, "answer": "No", "target": "broom", "split": "validation"}
{"id": "scene0117_00_5", "scene_id": "scene0117_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. kitchen cabinets\nB. cup\nC. toaster\nD. cups", "choices": ["A. kitchen cabinets", "B. cup", "C. toaster", "D. cups"], "answer": "A. kitchen cabinets", "target": "paper towel roll", "split": "validation"}
{"id": "scene0117_00_6", "scene_id": "scene0117_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 to the side of and above the kitchen cabinets, about 18 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 below the cups, about 33 cm away.", "target": "paper towel roll", "split": "validation"}
{"id": "scene0117_00_7", "scene_id": "scene0117_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\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen cabinets\", \"relation\": \"side_above\", \"distance_cm\": 18.2}, {\"object\": \"cup\", \"relation\": \"side_below\", \"distance_cm\": 28.3}]}", "target": "paper towel roll", "split": "validation"}
{"id": "scene0117_00_8", "scene_id": "scene0117_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": "validation"}
{"id": "scene0117_00_9", "scene_id": "scene0117_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. chair\nB. backpack\nC. sink\nD. paper towel dispenser", "choices": ["A. chair", "B. backpack", "C. sink", "D. paper towel dispenser"], "answer": "D. paper towel dispenser", "target": "soap dispenser", "split": "validation"}
{"id": "scene0117_00_10", "scene_id": "scene0117_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 below the paper towel dispenser, about 63 cm away. It is also to the side of and above the sink, about 65 cm away.", "target": "soap dispenser", "split": "validation"}
{"id": "scene0117_00_11", "scene_id": "scene0117_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\": \"paper towel dispenser\", \"references\": [{\"object\": \"paper towel dispenser\", \"relation\": \"side_below\", \"distance_cm\": 62.6}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 64.9}]}", "target": "soap dispenser", "split": "validation"}
{"id": "scene0117_00_12", "scene_id": "scene0117_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 toaster appear?", "choices": null, "answer": "Yes", "target": "toaster", "split": "validation"}
{"id": "scene0117_00_13", "scene_id": "scene0117_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 toaster in this scene?\nA. refrigerator\nB. cup\nC. cups\nD. paper towel roll", "choices": ["A. refrigerator", "B. cup", "C. cups", "D. paper towel roll"], "answer": "C. cups", "target": "toaster", "split": "validation"}
{"id": "scene0117_00_14", "scene_id": "scene0117_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 toaster in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toaster is on the shelf. It is also to the side of and below the cups, about 29 cm away. It is further near the refrigerator, about 36 cm away.", "target": "toaster", "split": "validation"}
{"id": "scene0117_00_15", "scene_id": "scene0117_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 toaster in structured JSON format.", "choices": null, "answer": "{\"target\": \"toaster\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"cups\", \"relation\": \"side_below\", \"distance_cm\": 28.9}, {\"object\": \"refrigerator\", \"relation\": \"near\", \"distance_cm\": 36.4}]}", "target": "toaster", "split": "validation"}
{"id": "scene0121_00_0", "scene_id": "scene0121_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": "validation"}
{"id": "scene0121_00_1", "scene_id": "scene0121_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. clothes\nB. mirror\nC. chair\nD. monitor", "choices": ["A. clothes", "B. mirror", "C. chair", "D. monitor"], "answer": "B. mirror", "target": "purse", "split": "validation"}
{"id": "scene0121_00_2", "scene_id": "scene0121_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 to the side of and above the mirror, about 15 cm away.", "target": "purse", "split": "validation"}
{"id": "scene0121_00_3", "scene_id": "scene0121_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\": \"clothes\", \"references\": [{\"object\": \"mirror\", \"relation\": \"side_above\", \"distance_cm\": 14.6}]}", "target": "purse", "split": "validation"}
{"id": "scene0136_00_0", "scene_id": "scene0136_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": "validation"}
{"id": "scene0136_00_1", "scene_id": "scene0136_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. paper\nC. cabinet\nD. box", "choices": ["A. cabinets", "B. paper", "C. cabinet", "D. box"], "answer": "B. paper", "target": "paper cutter", "split": "validation"}
{"id": "scene0136_00_2", "scene_id": "scene0136_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 under the paper, about 5 cm away. It is also to the side of and below the cabinet, about 48 cm away. It is further to the side of and above the box, about 56 cm away.", "target": "paper cutter", "split": "validation"}
{"id": "scene0136_00_3", "scene_id": "scene0136_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\": \"paper\", \"relation\": \"under\", \"distance_cm\": 4.6}, {\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 47.8}]}", "target": "paper cutter", "split": "validation"}
{"id": "scene0140_00_0", "scene_id": "scene0140_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": "validation"}
{"id": "scene0156_00_0", "scene_id": "scene0156_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": "validation"}
{"id": "scene0156_00_1", "scene_id": "scene0156_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. chair", "choices": ["A. desk", "B. keyboard", "C. monitor", "D. chair"], "answer": "D. chair", "target": "backpack", "split": "validation"}
{"id": "scene0156_00_2", "scene_id": "scene0156_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 51 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0156_00_3", "scene_id": "scene0156_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\": 50.7}]}", "target": "backpack", "split": "validation"}
{"id": "scene0164_00_0", "scene_id": "scene0164_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": "validation"}
{"id": "scene0164_00_1", "scene_id": "scene0164_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. beer bottles\nB. bag\nC. box\nD. card", "choices": ["A. beer bottles", "B. bag", "C. box", "D. card"], "answer": "B. bag", "target": "backpack", "split": "validation"}
{"id": "scene0164_00_2", "scene_id": "scene0164_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 bag, about 57 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0164_00_3", "scene_id": "scene0164_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\": \"bag\", \"references\": [{\"object\": \"bag\", \"relation\": \"near\", \"distance_cm\": 57.3}]}", "target": "backpack", "split": "validation"}
{"id": "scene0164_00_4", "scene_id": "scene0164_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": "validation"}
{"id": "scene0164_00_5", "scene_id": "scene0164_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. box\nB. kitchen cabinet\nC. backpack\nD. beer bottles", "choices": ["A. box", "B. kitchen cabinet", "C. backpack", "D. beer bottles"], "answer": "B. kitchen cabinet", "target": "bag", "split": "validation"}
{"id": "scene0164_00_6", "scene_id": "scene0164_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 kitchen cabinet, about 41 cm away. It is also near the backpack, about 57 cm away.", "target": "bag", "split": "validation"}
{"id": "scene0164_00_7", "scene_id": "scene0164_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\": \"kitchen cabinet\", \"references\": [{\"object\": \"kitchen cabinet\", \"relation\": \"side_below\", \"distance_cm\": 40.9}, {\"object\": \"backpack\", \"relation\": \"near\", \"distance_cm\": 57.3}]}", "target": "bag", "split": "validation"}
{"id": "scene0164_00_8", "scene_id": "scene0164_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 coffee kettle appear?", "choices": null, "answer": "Yes", "target": "coffee kettle", "split": "validation"}
{"id": "scene0164_00_9", "scene_id": "scene0164_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 coffee kettle in this scene?\nA. dishwasher\nB. picture\nC. dish rack\nD. faucet", "choices": ["A. dishwasher", "B. picture", "C. dish rack", "D. faucet"], "answer": "D. faucet", "target": "coffee kettle", "split": "validation"}
{"id": "scene0164_00_10", "scene_id": "scene0164_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 coffee kettle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The coffee kettle is on the counter. It is also to the side of and above the faucet, about 4 cm away. It is further to the side of and above the dishwasher, about 36 cm away.", "target": "coffee kettle", "split": "validation"}
{"id": "scene0164_00_11", "scene_id": "scene0164_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 coffee kettle in structured JSON format.", "choices": null, "answer": "{\"target\": \"coffee kettle\", \"support_surface\": \"counter\", \"references\": [{\"object\": \"faucet\", \"relation\": \"side_above\", \"distance_cm\": 3.6}, {\"object\": \"dishwasher\", \"relation\": \"side_above\", \"distance_cm\": 36.2}]}", "target": "coffee kettle", "split": "validation"}
{"id": "scene0164_00_12", "scene_id": "scene0164_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 hair dryer appear?", "choices": null, "answer": "No", "target": "hair dryer", "split": "validation"}
{"id": "scene0164_00_13", "scene_id": "scene0164_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 towel roll in this scene?\nA. soda stream\nB. beer bottles\nC. kitchen cabinets\nD. kitchen cabinet", "choices": ["A. soda stream", "B. beer bottles", "C. kitchen cabinets", "D. kitchen cabinet"], "answer": "D. kitchen cabinet", "target": "paper towel roll", "split": "validation"}
{"id": "scene0164_00_14", "scene_id": "scene0164_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 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 cabinet, about 12 cm away. It is also to the side of and above the kitchen cabinets, about 18 cm away. It is further near the soda stream, about 20 cm away.", "target": "paper towel roll", "split": "validation"}
{"id": "scene0164_00_15", "scene_id": "scene0164_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 towel roll in structured JSON format.", "choices": null, "answer": "{\"target\": \"paper towel roll\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"kitchen cabinet\", \"relation\": \"side_below\", \"distance_cm\": 12.5}, {\"object\": \"kitchen cabinets\", \"relation\": \"side_above\", \"distance_cm\": 18.0}]}", "target": "paper towel roll", "split": "validation"}
{"id": "scene0164_00_16", "scene_id": "scene0164_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 toaster appear?", "choices": null, "answer": "Yes", "target": "toaster", "split": "validation"}
{"id": "scene0164_00_17", "scene_id": "scene0164_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 toaster in this scene?\nA. microwave\nB. open kitchen cabinet\nC. kitchen cabinet\nD. box", "choices": ["A. microwave", "B. open kitchen cabinet", "C. kitchen cabinet", "D. box"], "answer": "A. microwave", "target": "toaster", "split": "validation"}
{"id": "scene0164_00_18", "scene_id": "scene0164_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 toaster in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The toaster is next to the microwave, about 4 cm away. It is also to the side of and below the open kitchen cabinet, about 16 cm away. It is further to the side of and below the kitchen cabinet, about 23 cm away.", "target": "toaster", "split": "validation"}
{"id": "scene0164_00_19", "scene_id": "scene0164_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 toaster in structured JSON format.", "choices": null, "answer": "{\"target\": \"toaster\", \"support_surface\": \"open kitchen cabinet\", \"references\": [{\"object\": \"microwave\", \"relation\": \"next_to\", \"distance_cm\": 4.5}, {\"object\": \"open kitchen cabinet\", \"relation\": \"side_below\", \"distance_cm\": 16.4}]}", "target": "toaster", "split": "validation"}
{"id": "scene0164_00_20", "scene_id": "scene0164_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 telephone appear?", "choices": null, "answer": "No", "target": "telephone", "split": "validation"}
{"id": "scene0164_00_21", "scene_id": "scene0164_00", "source_dataset": "scannet200", "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 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 refrigerator, about 6 cm away. It is also next to the carton, about 6 cm away. It is further to the side of and above the kitchen cabinets, about 12 cm away.", "target": "water bottle", "split": "validation"}
{"id": "scene0164_00_22", "scene_id": "scene0164_00", "source_dataset": "scannet200", "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 water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"kitchen cabinet\", \"references\": [{\"object\": \"refrigerator\", \"relation\": \"side_above\", \"distance_cm\": 5.8}, {\"object\": \"carton\", \"relation\": \"next_to\", \"distance_cm\": 5.8}]}", "target": "water bottle", "split": "validation"}
{"id": "scene0166_00_0", "scene_id": "scene0166_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": "Yes", "target": "ball", "split": "validation"}
{"id": "scene0166_00_1", "scene_id": "scene0166_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. bag\nB. book\nC. end table\nD. stool", "choices": ["A. bag", "B. book", "C. end table", "D. stool"], "answer": "D. stool", "target": "ball", "split": "validation"}
{"id": "scene0166_00_2", "scene_id": "scene0166_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 stool, about 10 cm away.", "target": "ball", "split": "validation"}
{"id": "scene0166_00_3", "scene_id": "scene0166_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\": \"stool\", \"references\": [{\"object\": \"stool\", \"relation\": \"side_below\", \"distance_cm\": 9.6}]}", "target": "ball", "split": "validation"}
{"id": "scene0166_00_4", "scene_id": "scene0166_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": "validation"}
{"id": "scene0166_00_5", "scene_id": "scene0166_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. thermostat\nB. end table\nC. pillow\nD. bag", "choices": ["A. thermostat", "B. end table", "C. pillow", "D. bag"], "answer": "B. end table", "target": "book", "split": "validation"}
{"id": "scene0166_00_6", "scene_id": "scene0166_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 bag, about 31 cm away. It is also to the side of and below the pillow, about 67 cm away. It is further to the side of and below the thermostat, about 72 cm away.", "target": "book", "split": "validation"}
{"id": "scene0166_00_7", "scene_id": "scene0166_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\": \"end table\", \"references\": [{\"object\": \"bag\", \"relation\": \"near\", \"distance_cm\": 31.1}, {\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 67.0}]}", "target": "book", "split": "validation"}
{"id": "scene0188_00_0", "scene_id": "scene0188_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": "validation"}
{"id": "scene0188_00_1", "scene_id": "scene0188_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\": \"cabinet\", \"references\": []}", "target": "backpack", "split": "validation"}
{"id": "scene0191_00_0", "scene_id": "scene0191_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": "validation"}
{"id": "scene0191_00_1", "scene_id": "scene0191_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. table", "choices": ["A. chair", "B. desk", "C. monitor", "D. table"], "answer": "A. chair", "target": "backpack", "split": "validation"}
{"id": "scene0191_00_2", "scene_id": "scene0191_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 35 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0191_00_3", "scene_id": "scene0191_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\": 35.0}]}", "target": "backpack", "split": "validation"}
{"id": "scene0196_00_0", "scene_id": "scene0196_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": "validation"}
{"id": "scene0196_00_1", "scene_id": "scene0196_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. monitor\nD. chair", "choices": ["A. table", "B. desk", "C. monitor", "D. chair"], "answer": "D. chair", "target": "backpack", "split": "validation"}
{"id": "scene0196_00_2", "scene_id": "scene0196_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 33 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0196_00_3", "scene_id": "scene0196_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\": 33.2}]}", "target": "backpack", "split": "validation"}
{"id": "scene0207_00_0", "scene_id": "scene0207_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": "validation"}
{"id": "scene0207_00_1", "scene_id": "scene0207_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. nightstand\nC. plate\nD. monitor", "choices": ["A. desk", "B. nightstand", "C. plate", "D. monitor"], "answer": "C. plate", "target": "paper towel roll", "split": "validation"}
{"id": "scene0207_00_2", "scene_id": "scene0207_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 plate, about 15 cm away. It is further to the side of and below the monitor, about 50 cm away.", "target": "paper towel roll", "split": "validation"}
{"id": "scene0207_00_3", "scene_id": "scene0207_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\": \"plate\", \"relation\": \"near\", \"distance_cm\": 15.1}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 50.1}]}", "target": "paper towel roll", "split": "validation"}
{"id": "scene0207_00_4", "scene_id": "scene0207_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": "validation"}
{"id": "scene0207_00_5", "scene_id": "scene0207_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. monitor\nB. nightstand\nC. desk\nD. paper towel roll", "choices": ["A. monitor", "B. nightstand", "C. desk", "D. paper towel roll"], "answer": "D. paper towel roll", "target": "plate", "split": "validation"}
{"id": "scene0207_00_6", "scene_id": "scene0207_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 on the desk. It is also near the paper towel roll, about 15 cm away. It is further to the side of and below the monitor, about 25 cm away.", "target": "plate", "split": "validation"}
{"id": "scene0207_00_7", "scene_id": "scene0207_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\": \"desk\", \"references\": [{\"object\": \"paper towel roll\", \"relation\": \"near\", \"distance_cm\": 15.1}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 25.1}]}", "target": "plate", "split": "validation"}
{"id": "scene0207_00_8", "scene_id": "scene0207_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": "validation"}
{"id": "scene0207_00_9", "scene_id": "scene0207_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. shoes\nB. nightstand\nC. monitor\nD. desk", "choices": ["A. shoes", "B. nightstand", "C. monitor", "D. desk"], "answer": "C. monitor", "target": "tissue box", "split": "validation"}
{"id": "scene0207_00_10", "scene_id": "scene0207_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 desk. It is also to the side of and below the monitor, about 26 cm away. It is further to the side of and above the nightstand, about 49 cm away.", "target": "tissue box", "split": "validation"}
{"id": "scene0207_00_11", "scene_id": "scene0207_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\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 26.1}, {\"object\": \"nightstand\", \"relation\": \"side_above\", \"distance_cm\": 49.4}]}", "target": "tissue box", "split": "validation"}
{"id": "scene0211_00_0", "scene_id": "scene0211_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": "validation"}
{"id": "scene0212_00_0", "scene_id": "scene0212_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": "validation"}
{"id": "scene0212_00_1", "scene_id": "scene0212_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. banister\nD. staircase", "choices": ["A. chair", "B. monitor", "C. banister", "D. staircase"], "answer": "D. staircase", "target": "backpack", "split": "validation"}
{"id": "scene0212_00_2", "scene_id": "scene0212_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 staircase, about 18 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0212_00_3", "scene_id": "scene0212_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\": \"banister\", \"references\": [{\"object\": \"staircase\", \"relation\": \"side_below\", \"distance_cm\": 17.9}]}", "target": "backpack", "split": "validation"}
{"id": "scene0213_00_0", "scene_id": "scene0213_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": "validation"}
{"id": "scene0213_00_1", "scene_id": "scene0213_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. armchair\nC. bag\nD. bench", "choices": ["A. coffee table", "B. armchair", "C. bag", "D. bench"], "answer": "D. bench", "target": "backpack", "split": "validation"}
{"id": "scene0213_00_2", "scene_id": "scene0213_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 bench, about 26 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0213_00_3", "scene_id": "scene0213_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\": \"coffee table\", \"references\": [{\"object\": \"bench\", \"relation\": \"side_below\", \"distance_cm\": 26.4}]}", "target": "backpack", "split": "validation"}
{"id": "scene0213_00_4", "scene_id": "scene0213_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": "validation"}
{"id": "scene0213_00_5", "scene_id": "scene0213_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. shelf\nB. armchair\nC. windowsill\nD. lamp", "choices": ["A. shelf", "B. armchair", "C. windowsill", "D. lamp"], "answer": "D. lamp", "target": "bag", "split": "validation"}
{"id": "scene0213_00_6", "scene_id": "scene0213_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 under the lamp. It is also to the side of and below the armchair, about 14 cm away. It is further to the side of and below the windowsill, about 34 cm away.", "target": "bag", "split": "validation"}
{"id": "scene0213_00_7", "scene_id": "scene0213_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\": \"shelf\", \"references\": [{\"object\": \"lamp\", \"relation\": \"under\", \"distance_cm\": 2.1}, {\"object\": \"armchair\", \"relation\": \"side_below\", \"distance_cm\": 14.0}]}", "target": "bag", "split": "validation"}
{"id": "scene0220_00_0", "scene_id": "scene0220_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": "validation"}
{"id": "scene0222_00_0", "scene_id": "scene0222_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": "validation"}
{"id": "scene0222_00_1", "scene_id": "scene0222_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. shoes\nC. chair\nD. monitor", "choices": ["A. desk", "B. shoes", "C. chair", "D. monitor"], "answer": "B. shoes", "target": "backpack", "split": "validation"}
{"id": "scene0222_00_2", "scene_id": "scene0222_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 25 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0222_00_3", "scene_id": "scene0222_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\": \"shoes\", \"relation\": \"side_above\", \"distance_cm\": 25.0}]}", "target": "backpack", "split": "validation"}
{"id": "scene0225_00_0", "scene_id": "scene0225_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": "validation"}
{"id": "scene0225_00_1", "scene_id": "scene0225_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. monitor\nC. chair\nD. trash can", "choices": ["A. cabinet", "B. monitor", "C. chair", "D. trash can"], "answer": "D. trash can", "target": "cup", "split": "validation"}
{"id": "scene0225_00_2", "scene_id": "scene0225_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 cabinet. It is also to the side of and above the trash can, about 17 cm away.", "target": "cup", "split": "validation"}
{"id": "scene0225_00_3", "scene_id": "scene0225_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\": \"cabinet\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 17.4}]}", "target": "cup", "split": "validation"}
{"id": "scene0226_00_0", "scene_id": "scene0226_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": "validation"}
{"id": "scene0226_00_1", "scene_id": "scene0226_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. sofa chair\nC. lamp\nD. table", "choices": ["A. monitor", "B. sofa chair", "C. lamp", "D. table"], "answer": "B. sofa chair", "target": "book", "split": "validation"}
{"id": "scene0226_00_2", "scene_id": "scene0226_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 above the sofa chair, about 17 cm away. It is further to the side of and below the lamp, about 63 cm away.", "target": "book", "split": "validation"}
{"id": "scene0226_00_3", "scene_id": "scene0226_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\": \"sofa chair\", \"relation\": \"side_above\", \"distance_cm\": 16.6}, {\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 63.0}]}", "target": "book", "split": "validation"}
{"id": "scene0233_00_0", "scene_id": "scene0233_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": "validation"}
{"id": "scene0233_00_1", "scene_id": "scene0233_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. bag\nC. cabinet\nD. chair", "choices": ["A. trash can", "B. bag", "C. cabinet", "D. chair"], "answer": "A. trash can", "target": "backpack", "split": "validation"}
{"id": "scene0233_00_2", "scene_id": "scene0233_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 trash can, about 17 cm away. It is also to the side of and below the cabinet, about 42 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0233_00_3", "scene_id": "scene0233_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\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 16.9}, {\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 41.8}]}", "target": "backpack", "split": "validation"}
{"id": "scene0252_00_0", "scene_id": "scene0252_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": "validation"}
{"id": "scene0252_00_1", "scene_id": "scene0252_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. monitor\nC. desk\nD. chair", "choices": ["A. table", "B. monitor", "C. desk", "D. chair"], "answer": "D. chair", "target": "backpack", "split": "validation"}
{"id": "scene0252_00_2", "scene_id": "scene0252_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 table, about 28 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0252_00_3", "scene_id": "scene0252_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_above\", \"distance_cm\": 27.6}]}", "target": "backpack", "split": "validation"}
{"id": "scene0255_00_0", "scene_id": "scene0255_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": "validation"}
{"id": "scene0255_00_1", "scene_id": "scene0255_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. bucket\nB. printer\nC. cabinet\nD. paper cutter", "choices": ["A. bucket", "B. printer", "C. cabinet", "D. paper cutter"], "answer": "D. paper cutter", "target": "bottle", "split": "validation"}
{"id": "scene0255_00_2", "scene_id": "scene0255_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 paper cutter, about 6 cm away. It is further to the side of and below the printer, about 25 cm away.", "target": "bottle", "split": "validation"}
{"id": "scene0255_00_3", "scene_id": "scene0255_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\": \"paper cutter\", \"relation\": \"side_above\", \"distance_cm\": 5.7}, {\"object\": \"printer\", \"relation\": \"side_below\", \"distance_cm\": 25.0}]}", "target": "bottle", "split": "validation"}
{"id": "scene0255_00_4", "scene_id": "scene0255_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": "validation"}
{"id": "scene0255_00_5", "scene_id": "scene0255_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. bottle\nB. cabinet\nC. printer\nD. bucket", "choices": ["A. bottle", "B. cabinet", "C. printer", "D. bucket"], "answer": "A. bottle", "target": "paper cutter", "split": "validation"}
{"id": "scene0255_00_6", "scene_id": "scene0255_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 bottle, about 6 cm away. It is also to the side of and above the cabinet, about 11 cm away. It is further to the side of and below the bucket, about 21 cm away.", "target": "paper cutter", "split": "validation"}
{"id": "scene0255_00_7", "scene_id": "scene0255_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\": \"cabinet\", \"references\": [{\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 5.7}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 10.6}]}", "target": "paper cutter", "split": "validation"}
{"id": "scene0262_00_0", "scene_id": "scene0262_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": "validation"}
{"id": "scene0262_00_1", "scene_id": "scene0262_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. bottle\nB. clock\nC. pillow\nD. book", "choices": ["A. bottle", "B. clock", "C. pillow", "D. book"], "answer": "C. pillow", "target": "backpack", "split": "validation"}
{"id": "scene0262_00_2", "scene_id": "scene0262_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 22 cm away. It is also to the side of and below the book, about 71 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0262_00_3", "scene_id": "scene0262_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\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 21.7}, {\"object\": \"book\", \"relation\": \"side_below\", \"distance_cm\": 70.8}]}", "target": "backpack", "split": "validation"}
{"id": "scene0262_00_4", "scene_id": "scene0262_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": "validation"}
{"id": "scene0262_00_5", "scene_id": "scene0262_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. plant\nB. clock\nC. bottle\nD. backpack", "choices": ["A. plant", "B. clock", "C. bottle", "D. backpack"], "answer": "A. plant", "target": "book", "split": "validation"}
{"id": "scene0262_00_6", "scene_id": "scene0262_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 plant, about 27 cm away. It is also to the side of and above the backpack, about 71 cm away.", "target": "book", "split": "validation"}
{"id": "scene0262_00_7", "scene_id": "scene0262_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\": \"backpack\", \"references\": [{\"object\": \"plant\", \"relation\": \"side_above\", \"distance_cm\": 26.6}, {\"object\": \"backpack\", \"relation\": \"side_above\", \"distance_cm\": 70.8}]}", "target": "book", "split": "validation"}
{"id": "scene0262_00_8", "scene_id": "scene0262_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": "validation"}
{"id": "scene0262_00_9", "scene_id": "scene0262_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 bottle in this scene?\nA. office chair\nB. backpack\nC. telephone\nD. book", "choices": ["A. office chair", "B. backpack", "C. telephone", "D. book"], "answer": "C. telephone", "target": "bottle", "split": "validation"}
{"id": "scene0262_00_10", "scene_id": "scene0262_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 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 18 cm away. It is also to the side of and above the office chair, about 54 cm away.", "target": "bottle", "split": "validation"}
{"id": "scene0262_00_11", "scene_id": "scene0262_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 bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"office chair\", \"references\": [{\"object\": \"telephone\", \"relation\": \"side_above\", \"distance_cm\": 18.0}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 54.4}]}", "target": "bottle", "split": "validation"}
{"id": "scene0262_00_12", "scene_id": "scene0262_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 broom appear?", "choices": null, "answer": "No", "target": "broom", "split": "validation"}
{"id": "scene0262_00_13", "scene_id": "scene0262_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 telephone appear?", "choices": null, "answer": "Yes", "target": "telephone", "split": "validation"}
{"id": "scene0262_00_14", "scene_id": "scene0262_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. bottle\nB. backpack\nC. book\nD. office chair", "choices": ["A. bottle", "B. backpack", "C. book", "D. office chair"], "answer": "A. bottle", "target": "telephone", "split": "validation"}
{"id": "scene0262_00_15", "scene_id": "scene0262_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 bottle, about 18 cm away. It is further to the side of and above the office chair, about 56 cm away.", "target": "telephone", "split": "validation"}
{"id": "scene0262_00_16", "scene_id": "scene0262_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\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 18.0}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 55.5}]}", "target": "telephone", "split": "validation"}
{"id": "scene0264_00_0", "scene_id": "scene0264_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": "validation"}
{"id": "scene0264_00_1", "scene_id": "scene0264_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. board\nB. bulletin board\nC. monitor\nD. books", "choices": ["A. board", "B. bulletin board", "C. monitor", "D. books"], "answer": "C. monitor", "target": "book", "split": "validation"}
{"id": "scene0264_00_2", "scene_id": "scene0264_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 18 cm away.", "target": "book", "split": "validation"}
{"id": "scene0264_00_3", "scene_id": "scene0264_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\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 18.5}]}", "target": "book", "split": "validation"}
{"id": "scene0264_00_4", "scene_id": "scene0264_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": "validation"}
{"id": "scene0264_00_5", "scene_id": "scene0264_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. printer\nB. computer tower\nC. monitor\nD. bulletin board", "choices": ["A. printer", "B. computer tower", "C. monitor", "D. bulletin board"], "answer": "C. monitor", "target": "keyboard", "split": "validation"}
{"id": "scene0264_00_6", "scene_id": "scene0264_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 monitor, about 12 cm away. It is also to the side of and below the bulletin board, about 38 cm away. It is further to the side of and above the computer tower, about 43 cm away.", "target": "keyboard", "split": "validation"}
{"id": "scene0264_00_7", "scene_id": "scene0264_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\": \"office chair\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 12.2}, {\"object\": \"bulletin board\", \"relation\": \"side_below\", \"distance_cm\": 37.9}]}", "target": "keyboard", "split": "validation"}
{"id": "scene0264_00_8", "scene_id": "scene0264_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": "Yes", "target": "tray", "split": "validation"}
{"id": "scene0264_00_9", "scene_id": "scene0264_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. monitor\nB. office chair\nC. picture\nD. bulletin board", "choices": ["A. monitor", "B. office chair", "C. picture", "D. bulletin board"], "answer": "A. monitor", "target": "tray", "split": "validation"}
{"id": "scene0264_00_10", "scene_id": "scene0264_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 to the side of and below the monitor, about 26 cm away. It is also to the side of and above the office chair, about 41 cm away. It is further to the side of and below the bulletin board, about 42 cm away.", "target": "tray", "split": "validation"}
{"id": "scene0264_00_11", "scene_id": "scene0264_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\": \"office chair\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 25.7}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 41.4}]}", "target": "tray", "split": "validation"}
{"id": "scene0274_00_0", "scene_id": "scene0274_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": "validation"}
{"id": "scene0274_00_1", "scene_id": "scene0274_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. table\nB. monitor\nC. box\nD. recycling bin", "choices": ["A. table", "B. monitor", "C. box", "D. recycling bin"], "answer": "C. box", "target": "telephone", "split": "validation"}
{"id": "scene0274_00_2", "scene_id": "scene0274_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 to the side of and below the box, about 17 cm away. It is further to the side of and below the monitor, about 31 cm away.", "target": "telephone", "split": "validation"}
{"id": "scene0274_00_3", "scene_id": "scene0274_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\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 17.5}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 30.8}]}", "target": "telephone", "split": "validation"}
{"id": "scene0278_00_0", "scene_id": "scene0278_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": "validation"}
{"id": "scene0278_00_1", "scene_id": "scene0278_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. counter\nB. cabinets\nC. trash can\nD. copier", "choices": ["A. counter", "B. cabinets", "C. trash can", "D. copier"], "answer": "D. copier", "target": "broom", "split": "validation"}
{"id": "scene0278_00_2", "scene_id": "scene0278_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 copier, about 42 cm away. It is also to the side of and above the trash can, about 54 cm away.", "target": "broom", "split": "validation"}
{"id": "scene0278_00_3", "scene_id": "scene0278_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\": \"trash can\", \"references\": [{\"object\": \"copier\", \"relation\": \"side_below\", \"distance_cm\": 42.1}, {\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 53.8}]}", "target": "broom", "split": "validation"}
{"id": "scene0278_00_4", "scene_id": "scene0278_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": "validation"}
{"id": "scene0278_00_5", "scene_id": "scene0278_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. broom\nB. copier\nC. counter\nD. cabinets", "choices": ["A. broom", "B. copier", "C. counter", "D. cabinets"], "answer": "D. cabinets", "target": "paper cutter", "split": "validation"}
{"id": "scene0278_00_6", "scene_id": "scene0278_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 on the counter. It is also to the side of and above the cabinets, about 8 cm away.", "target": "paper cutter", "split": "validation"}
{"id": "scene0278_00_7", "scene_id": "scene0278_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\": \"cabinets\", \"relation\": \"side_above\", \"distance_cm\": 8.5}]}", "target": "paper cutter", "split": "validation"}
{"id": "scene0288_00_0", "scene_id": "scene0288_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": "validation"}
{"id": "scene0288_00_1", "scene_id": "scene0288_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. desk\nC. monitor\nD. backpack", "choices": ["A. chair", "B. desk", "C. monitor", "D. backpack"], "answer": "A. chair", "target": "telephone", "split": "validation"}
{"id": "scene0288_00_2", "scene_id": "scene0288_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 37 cm away.", "target": "telephone", "split": "validation"}
{"id": "scene0288_00_3", "scene_id": "scene0288_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\": 37.5}]}", "target": "telephone", "split": "validation"}
{"id": "scene0304_00_0", "scene_id": "scene0304_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": "validation"}
{"id": "scene0304_00_1", "scene_id": "scene0304_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\": \"cabinet\", \"references\": []}", "target": "backpack", "split": "validation"}
{"id": "scene0310_00_0", "scene_id": "scene0310_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": "Yes", "target": "plate", "split": "validation"}
{"id": "scene0310_00_1", "scene_id": "scene0310_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. box\nB. dish rack\nC. coffee maker\nD. cabinet", "choices": ["A. box", "B. dish rack", "C. coffee maker", "D. cabinet"], "answer": "A. box", "target": "plate", "split": "validation"}
{"id": "scene0310_00_2", "scene_id": "scene0310_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 to the side of and above the box, about 23 cm away. It is also near the dish rack, about 28 cm away. It is further to the side of and above the cabinet, about 48 cm away.", "target": "plate", "split": "validation"}
{"id": "scene0310_00_3", "scene_id": "scene0310_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\": \"dish rack\", \"references\": [{\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 22.9}, {\"object\": \"dish rack\", \"relation\": \"near\", \"distance_cm\": 27.6}]}", "target": "plate", "split": "validation"}
{"id": "scene0310_00_4", "scene_id": "scene0310_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 dispenser appear?", "choices": null, "answer": "Yes", "target": "soap dispenser", "split": "validation"}
{"id": "scene0310_00_5", "scene_id": "scene0310_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. dish rack\nD. kitchen cabinet", "choices": ["A. kitchen counter", "B. sink", "C. dish rack", "D. kitchen cabinet"], "answer": "D. kitchen cabinet", "target": "soap dispenser", "split": "validation"}
{"id": "scene0310_00_6", "scene_id": "scene0310_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 cabinet, about 24 cm away. It is also to the side of and above the kitchen counter, about 32 cm away. It is further to the side of and above the dish rack, about 34 cm away.", "target": "soap dispenser", "split": "validation"}
{"id": "scene0310_00_7", "scene_id": "scene0310_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\": \"dish rack\", \"references\": [{\"object\": \"kitchen cabinet\", \"relation\": \"side_below\", \"distance_cm\": 24.3}, {\"object\": \"kitchen counter\", \"relation\": \"side_above\", \"distance_cm\": 32.4}]}", "target": "soap dispenser", "split": "validation"}
{"id": "scene0310_00_8", "scene_id": "scene0310_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": "No", "target": "tissue box", "split": "validation"}
{"id": "scene0310_00_9", "scene_id": "scene0310_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. cabinet\nB. shelf\nC. coffee maker\nD. box", "choices": ["A. cabinet", "B. shelf", "C. coffee maker", "D. box"], "answer": "C. coffee maker", "target": "tray", "split": "validation"}
{"id": "scene0310_00_10", "scene_id": "scene0310_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 to the side of and below the coffee maker, about 26 cm away. It is also to the side of and below the shelf, about 48 cm away.", "target": "tray", "split": "validation"}
{"id": "scene0310_00_11", "scene_id": "scene0310_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\": \"kitchen cabinet\", \"references\": [{\"object\": \"coffee maker\", \"relation\": \"side_below\", \"distance_cm\": 26.2}, {\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 48.3}]}", "target": "tray", "split": "validation"}
{"id": "scene0310_00_12", "scene_id": "scene0310_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 backpack appear?", "choices": null, "answer": "No", "target": "backpack", "split": "validation"}
{"id": "scene0310_00_13", "scene_id": "scene0310_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 water pitcher in this scene?\nA. soap dispenser\nB. dish rack\nC. cabinet\nD. box", "choices": ["A. soap dispenser", "B. dish rack", "C. cabinet", "D. box"], "answer": "D. box", "target": "water pitcher", "split": "validation"}
{"id": "scene0310_00_14", "scene_id": "scene0310_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 water pitcher in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water pitcher is to the side of and above the box, about 26 cm away. It is also to the side of and above the dish rack, about 29 cm away. It is further to the side of and above the cabinet, about 51 cm away.", "target": "water pitcher", "split": "validation"}
{"id": "scene0310_00_15", "scene_id": "scene0310_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 water pitcher in structured JSON format.", "choices": null, "answer": "{\"target\": \"water pitcher\", \"support_surface\": \"dish rack\", \"references\": [{\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 25.9}, {\"object\": \"dish rack\", \"relation\": \"side_above\", \"distance_cm\": 28.6}]}", "target": "water pitcher", "split": "validation"}
{"id": "scene0327_00_0", "scene_id": "scene0327_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": "validation"}
{"id": "scene0327_00_1", "scene_id": "scene0327_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. kitchen cabinet\nC. range hood\nD. stove", "choices": ["A. sink", "B. kitchen cabinet", "C. range hood", "D. stove"], "answer": "D. stove", "target": "paper towel roll", "split": "validation"}
{"id": "scene0327_00_2", "scene_id": "scene0327_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 stove, about 10 cm away. It is also to the side of and below the kitchen cabinet, about 16 cm away. It is further to the side of and below the range hood, about 38 cm away.", "target": "paper towel roll", "split": "validation"}
{"id": "scene0327_00_3", "scene_id": "scene0327_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\": \"kitchen counter\", \"references\": [{\"object\": \"stove\", \"relation\": \"side_above\", \"distance_cm\": 10.1}, {\"object\": \"kitchen cabinet\", \"relation\": \"side_below\", \"distance_cm\": 16.5}]}", "target": "paper towel roll", "split": "validation"}
{"id": "scene0327_00_4", "scene_id": "scene0327_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": "validation"}
{"id": "scene0331_00_0", "scene_id": "scene0331_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": "validation"}
{"id": "scene0331_00_1", "scene_id": "scene0331_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. dresser\nB. recycling bin\nC. trash can\nD. shoes", "choices": ["A. dresser", "B. recycling bin", "C. trash can", "D. shoes"], "answer": "D. shoes", "target": "backpack", "split": "validation"}
{"id": "scene0331_00_2", "scene_id": "scene0331_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 3 cm away. It is also to the side of and below the trash can, about 11 cm away. It is further to the side of and below the recycling bin, about 31 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0331_00_3", "scene_id": "scene0331_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\": \"shoes\", \"relation\": \"side_above\", \"distance_cm\": 2.6}, {\"object\": \"trash can\", \"relation\": \"side_below\", \"distance_cm\": 10.8}]}", "target": "backpack", "split": "validation"}
{"id": "scene0331_00_4", "scene_id": "scene0331_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": "validation"}
{"id": "scene0331_00_5", "scene_id": "scene0331_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. plant\nB. cup\nC. clothes\nD. bookshelf", "choices": ["A. plant", "B. cup", "C. clothes", "D. bookshelf"], "answer": "D. bookshelf", "target": "bag", "split": "validation"}
{"id": "scene0331_00_6", "scene_id": "scene0331_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 bookshelf, about 5 cm away. It is also to the side of and above the cup, about 14 cm away. It is further to the side of and below the plant, about 16 cm away.", "target": "bag", "split": "validation"}
{"id": "scene0331_00_7", "scene_id": "scene0331_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\": \"bookshelf\", \"references\": [{\"object\": \"bookshelf\", \"relation\": \"side_above\", \"distance_cm\": 5.2}, {\"object\": \"cup\", \"relation\": \"side_above\", \"distance_cm\": 13.8}]}", "target": "bag", "split": "validation"}
{"id": "scene0331_00_8", "scene_id": "scene0331_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 book appear?", "choices": null, "answer": "Yes", "target": "book", "split": "validation"}
{"id": "scene0331_00_9", "scene_id": "scene0331_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. fan\nB. plant\nC. cup\nD. desk", "choices": ["A. fan", "B. plant", "C. cup", "D. desk"], "answer": "C. cup", "target": "book", "split": "validation"}
{"id": "scene0331_00_10", "scene_id": "scene0331_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 bookshelf. It is also to the side of and above the cup, about 9 cm away. It is further to the side of and above the fan, about 12 cm away.", "target": "book", "split": "validation"}
{"id": "scene0331_00_11", "scene_id": "scene0331_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\": \"bookshelf\", \"references\": [{\"object\": \"cup\", \"relation\": \"side_above\", \"distance_cm\": 9.2}, {\"object\": \"fan\", \"relation\": \"side_above\", \"distance_cm\": 12.5}]}", "target": "book", "split": "validation"}
{"id": "scene0331_00_12", "scene_id": "scene0331_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 ball appear?", "choices": null, "answer": "No", "target": "ball", "split": "validation"}
{"id": "scene0331_00_13", "scene_id": "scene0331_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. chair\nB. box\nC. block\nD. storage organizer", "choices": ["A. chair", "B. box", "C. block", "D. storage organizer"], "answer": "D. storage organizer", "target": "purse", "split": "validation"}
{"id": "scene0331_00_14", "scene_id": "scene0331_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 storage organizer, about 9 cm away. It is also to the side of and below the box, about 11 cm away. It is further next to the block, about 11 cm away.", "target": "purse", "split": "validation"}
{"id": "scene0331_00_15", "scene_id": "scene0331_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\": \"desk\", \"references\": [{\"object\": \"storage organizer\", \"relation\": \"side_below\", \"distance_cm\": 9.4}, {\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 10.6}]}", "target": "purse", "split": "validation"}
{"id": "scene0333_00_0", "scene_id": "scene0333_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": "validation"}
{"id": "scene0333_00_1", "scene_id": "scene0333_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. chair\nC. person\nD. monitor", "choices": ["A. computer tower", "B. chair", "C. person", "D. monitor"], "answer": "D. monitor", "target": "laptop", "split": "validation"}
{"id": "scene0333_00_2", "scene_id": "scene0333_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 under the monitor, about 3 cm away. It is further to the side of and above the person, about 11 cm away.", "target": "laptop", "split": "validation"}
{"id": "scene0333_00_3", "scene_id": "scene0333_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\": \"under\", \"distance_cm\": 3.1}, {\"object\": \"person\", \"relation\": \"side_above\", \"distance_cm\": 11.2}]}", "target": "laptop", "split": "validation"}
{"id": "scene0336_00_0", "scene_id": "scene0336_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": "validation"}
{"id": "scene0336_00_1", "scene_id": "scene0336_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. books\nC. step stool\nD. picture", "choices": ["A. chair", "B. books", "C. step stool", "D. picture"], "answer": "C. step stool", "target": "book", "split": "validation"}
{"id": "scene0336_00_2", "scene_id": "scene0336_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 above the step stool, about 16 cm away. It is also to the side of and below the picture, about 62 cm away.", "target": "book", "split": "validation"}
{"id": "scene0336_00_3", "scene_id": "scene0336_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\": \"step stool\", \"references\": [{\"object\": \"step stool\", \"relation\": \"side_above\", \"distance_cm\": 16.1}, {\"object\": \"picture\", \"relation\": \"side_below\", \"distance_cm\": 61.9}]}", "target": "book", "split": "validation"}
{"id": "scene0336_00_4", "scene_id": "scene0336_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": "validation"}
{"id": "scene0336_00_5", "scene_id": "scene0336_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. book\nB. monitor\nC. desk\nD. chair", "choices": ["A. book", "B. monitor", "C. desk", "D. chair"], "answer": "C. desk", "target": "tissue box", "split": "validation"}
{"id": "scene0336_00_6", "scene_id": "scene0336_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 on the desk. It is also to the side of and above the chair, about 28 cm away. It is further to the side of and below the monitor, about 39 cm away.", "target": "tissue box", "split": "validation"}
{"id": "scene0336_00_7", "scene_id": "scene0336_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\": \"desk\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 28.2}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 39.4}]}", "target": "tissue box", "split": "validation"}
{"id": "scene0342_00_0", "scene_id": "scene0342_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": "validation"}
{"id": "scene0342_00_1", "scene_id": "scene0342_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": "validation"}
{"id": "scene0347_00_0", "scene_id": "scene0347_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": "validation"}
{"id": "scene0347_00_1", "scene_id": "scene0347_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 coffee kettle in this scene?\nA. monitor\nB. kitchen counter\nC. coffee maker\nD. cup", "choices": ["A. monitor", "B. kitchen counter", "C. coffee maker", "D. cup"], "answer": "C. coffee maker", "target": "coffee kettle", "split": "validation"}
{"id": "scene0347_00_2", "scene_id": "scene0347_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 coffee kettle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The coffee kettle is to the side of and below the coffee maker, about 15 cm away. It is also to the side of and above the cup, about 47 cm away.", "target": "coffee kettle", "split": "validation"}
{"id": "scene0347_00_3", "scene_id": "scene0347_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 coffee kettle in structured JSON format.", "choices": null, "answer": "{\"target\": \"coffee kettle\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"coffee maker\", \"relation\": \"side_below\", \"distance_cm\": 14.8}, {\"object\": \"cup\", \"relation\": \"side_above\", \"distance_cm\": 47.1}]}", "target": "coffee kettle", "split": "validation"}
{"id": "scene0349_00_0", "scene_id": "scene0349_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": "validation"}
{"id": "scene0349_00_1", "scene_id": "scene0349_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 stuffed animal appear?", "choices": null, "answer": "No", "target": "stuffed animal", "split": "validation"}
{"id": "scene0349_00_2", "scene_id": "scene0349_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 broom in this scene?\nA. clothes dryer\nB. bag\nC. sink\nD. washing machine", "choices": ["A. clothes dryer", "B. bag", "C. sink", "D. washing machine"], "answer": "C. sink", "target": "broom", "split": "validation"}
{"id": "scene0349_00_3", "scene_id": "scene0349_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 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 sink, about 13 cm away. It is also to the side of and below the washing machine, about 39 cm away.", "target": "broom", "split": "validation"}
{"id": "scene0349_00_4", "scene_id": "scene0349_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 broom in structured JSON format.", "choices": null, "answer": "{\"target\": \"broom\", \"support_surface\": \"dustpan\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_below\", \"distance_cm\": 13.2}, {\"object\": \"washing machine\", \"relation\": \"side_below\", \"distance_cm\": 38.7}]}", "target": "broom", "split": "validation"}
{"id": "scene0349_00_5", "scene_id": "scene0349_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 shoe appear?", "choices": null, "answer": "No", "target": "shoe", "split": "validation"}
{"id": "scene0349_00_6", "scene_id": "scene0349_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 dustpan in this scene?\nA. washing machine\nB. sink\nC. bag\nD. trash can", "choices": ["A. washing machine", "B. sink", "C. bag", "D. trash can"], "answer": "D. trash can", "target": "dustpan", "split": "validation"}
{"id": "scene0349_00_7", "scene_id": "scene0349_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 dustpan in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The dustpan is next to the trash can, about 6 cm away. It is also to the side of and below the sink, about 12 cm away. It is further to the side of and below the washing machine, about 42 cm away.", "target": "dustpan", "split": "validation"}
{"id": "scene0349_00_8", "scene_id": "scene0349_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 dustpan in structured JSON format.", "choices": null, "answer": "{\"target\": \"dustpan\", \"support_surface\": \"broom\", \"references\": [{\"object\": \"trash can\", \"relation\": \"next_to\", \"distance_cm\": 5.6}, {\"object\": \"sink\", \"relation\": \"side_below\", \"distance_cm\": 12.5}]}", "target": "dustpan", "split": "validation"}
{"id": "scene0352_00_0", "scene_id": "scene0352_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": "validation"}
{"id": "scene0352_00_1", "scene_id": "scene0352_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. shelf\nC. cabinet\nD. pillow", "choices": ["A. computer tower", "B. shelf", "C. cabinet", "D. pillow"], "answer": "B. shelf", "target": "backpack", "split": "validation"}
{"id": "scene0352_00_2", "scene_id": "scene0352_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 to the side of and below the shelf, about 12 cm away. It is further to the side of and above the pillow, about 46 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0352_00_3", "scene_id": "scene0352_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\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 11.7}, {\"object\": \"pillow\", \"relation\": \"side_above\", \"distance_cm\": 45.6}]}", "target": "backpack", "split": "validation"}
{"id": "scene0352_00_4", "scene_id": "scene0352_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": "No", "target": "dumbbell", "split": "validation"}
{"id": "scene0352_00_5", "scene_id": "scene0352_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. backpack\nB. cabinet\nC. shelf\nD. computer tower", "choices": ["A. backpack", "B. cabinet", "C. shelf", "D. computer tower"], "answer": "C. shelf", "target": "telephone", "split": "validation"}
{"id": "scene0352_00_6", "scene_id": "scene0352_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 cabinet. It is also to the side of and above the shelf, about 24 cm away. It is further to the side of and above the computer tower, about 41 cm away.", "target": "telephone", "split": "validation"}
{"id": "scene0352_00_7", "scene_id": "scene0352_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\": \"cabinet\", \"references\": [{\"object\": \"shelf\", \"relation\": \"side_above\", \"distance_cm\": 23.8}, {\"object\": \"computer tower\", \"relation\": \"side_above\", \"distance_cm\": 40.6}]}", "target": "telephone", "split": "validation"}
{"id": "scene0353_00_0", "scene_id": "scene0353_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": "validation"}
{"id": "scene0353_00_1", "scene_id": "scene0353_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. chair\nB. blanket\nC. shelf\nD. laptop", "choices": ["A. chair", "B. blanket", "C. shelf", "D. laptop"], "answer": "D. laptop", "target": "bottle", "split": "validation"}
{"id": "scene0353_00_2", "scene_id": "scene0353_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 desk. It is also to the side of and above the laptop, about 20 cm away. It is further to the side of and above the chair, about 40 cm away.", "target": "bottle", "split": "validation"}
{"id": "scene0353_00_3", "scene_id": "scene0353_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\": \"desk\", \"references\": [{\"object\": \"laptop\", \"relation\": \"side_above\", \"distance_cm\": 20.5}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 39.9}]}", "target": "bottle", "split": "validation"}
{"id": "scene0353_00_4", "scene_id": "scene0353_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": "validation"}
{"id": "scene0353_00_5", "scene_id": "scene0353_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. jar\nB. book\nC. laptop\nD. shelf", "choices": ["A. jar", "B. book", "C. laptop", "D. shelf"], "answer": "A. jar", "target": "bowl", "split": "validation"}
{"id": "scene0353_00_6", "scene_id": "scene0353_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 desk. It is also under the jar, about 3 cm away. It is further next to the book, about 13 cm away.", "target": "bowl", "split": "validation"}
{"id": "scene0353_00_7", "scene_id": "scene0353_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\": \"desk\", \"references\": [{\"object\": \"jar\", \"relation\": \"under\", \"distance_cm\": 3.3}, {\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 12.7}]}", "target": "bowl", "split": "validation"}
{"id": "scene0353_00_8", "scene_id": "scene0353_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 bag appear?", "choices": null, "answer": "No", "target": "paper bag", "split": "validation"}
{"id": "scene0353_00_9", "scene_id": "scene0353_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. desk lamp\nB. chair\nC. book\nD. bottle", "choices": ["A. desk lamp", "B. chair", "C. book", "D. bottle"], "answer": "A. desk lamp", "target": "laptop", "split": "validation"}
{"id": "scene0353_00_10", "scene_id": "scene0353_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 below the desk lamp, about 19 cm away. It is also to the side of and below the bottle, about 20 cm away. It is further to the side of and above the chair, about 24 cm away.", "target": "laptop", "split": "validation"}
{"id": "scene0353_00_11", "scene_id": "scene0353_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\": \"desk lamp\", \"relation\": \"side_below\", \"distance_cm\": 19.2}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 20.5}]}", "target": "laptop", "split": "validation"}
{"id": "scene0353_00_12", "scene_id": "scene0353_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 dustpan appear?", "choices": null, "answer": "No", "target": "dustpan", "split": "validation"}
{"id": "scene0353_00_13", "scene_id": "scene0353_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 plate in this scene?\nA. book\nB. backpack\nC. shelf\nD. chair", "choices": ["A. book", "B. backpack", "C. shelf", "D. chair"], "answer": "A. book", "target": "plate", "split": "validation"}
{"id": "scene0353_00_14", "scene_id": "scene0353_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 plate in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The plate is near the book, about 31 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 below the shelf, about 43 cm away.", "target": "plate", "split": "validation"}
{"id": "scene0353_00_15", "scene_id": "scene0353_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 plate in structured JSON format.", "choices": null, "answer": "{\"target\": \"plate\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"book\", \"relation\": \"near\", \"distance_cm\": 31.4}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 34.2}]}", "target": "plate", "split": "validation"}
{"id": "scene0358_00_0", "scene_id": "scene0358_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": "validation"}
{"id": "scene0358_00_1", "scene_id": "scene0358_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. monitor\nC. chair\nD. keyboard", "choices": ["A. desk", "B. monitor", "C. chair", "D. keyboard"], "answer": "C. chair", "target": "backpack", "split": "validation"}
{"id": "scene0369_00_0", "scene_id": "scene0369_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": "validation"}
{"id": "scene0370_00_0", "scene_id": "scene0370_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": "validation"}
{"id": "scene0370_00_1", "scene_id": "scene0370_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. stand\nB. office chair\nC. recycling bin\nD. microwave", "choices": ["A. stand", "B. office chair", "C. recycling bin", "D. microwave"], "answer": "A. stand", "target": "cup", "split": "validation"}
{"id": "scene0370_00_2", "scene_id": "scene0370_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 under the stand, about 5 cm away. It is further to the side of and above the office chair, about 7 cm away.", "target": "cup", "split": "validation"}
{"id": "scene0370_00_3", "scene_id": "scene0370_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\": \"stand\", \"relation\": \"under\", \"distance_cm\": 5.0}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 7.1}]}", "target": "cup", "split": "validation"}
{"id": "scene0370_00_4", "scene_id": "scene0370_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": "Yes", "target": "water bottle", "split": "validation"}
{"id": "scene0370_00_5", "scene_id": "scene0370_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 bottle in this scene?\nA. table\nB. trash can\nC. recycling bin\nD. cup", "choices": ["A. table", "B. trash can", "C. recycling bin", "D. cup"], "answer": "A. table", "target": "water bottle", "split": "validation"}
{"id": "scene0370_00_6", "scene_id": "scene0370_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 bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is to the side of and below the table, about 5 cm away. It is also to the side of and below the recycling bin, about 14 cm away. It is further to the side of and below the trash can, about 50 cm away.", "target": "water bottle", "split": "validation"}
{"id": "scene0370_00_7", "scene_id": "scene0370_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 bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 5.0}, {\"object\": \"recycling bin\", \"relation\": \"side_below\", \"distance_cm\": 14.0}]}", "target": "water bottle", "split": "validation"}
{"id": "scene0404_00_0", "scene_id": "scene0404_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": "validation"}
{"id": "scene0404_00_1", "scene_id": "scene0404_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 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 swiffer, about 44 cm away. It is further to the side of and above the dustpan, about 44 cm away.", "target": "bottle", "split": "validation"}
{"id": "scene0404_00_2", "scene_id": "scene0404_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 bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"table\", \"references\": [{\"object\": \"swiffer\", \"relation\": \"side_above\", \"distance_cm\": 44.4}, {\"object\": \"dustpan\", \"relation\": \"side_above\", \"distance_cm\": 44.4}]}", "target": "bottle", "split": "validation"}
{"id": "scene0404_00_3", "scene_id": "scene0404_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 cd case appear?", "choices": null, "answer": "No", "target": "cd case", "split": "validation"}
{"id": "scene0404_00_4", "scene_id": "scene0404_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 dustpan in this scene?\nA. bottle\nB. storage bin\nC. table\nD. cloth", "choices": ["A. bottle", "B. storage bin", "C. table", "D. cloth"], "answer": "C. table", "target": "dustpan", "split": "validation"}
{"id": "scene0404_00_5", "scene_id": "scene0404_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 dustpan in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The dustpan is under the table, about 3 cm away. It is also to the side of and above the storage bin, about 28 cm away. It is further to the side of and below the bottle, about 44 cm away.", "target": "dustpan", "split": "validation"}
{"id": "scene0404_00_6", "scene_id": "scene0404_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 dustpan in structured JSON format.", "choices": null, "answer": "{\"target\": \"dustpan\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"under\", \"distance_cm\": 3.1}, {\"object\": \"storage bin\", \"relation\": \"side_above\", \"distance_cm\": 28.0}]}", "target": "dustpan", "split": "validation"}
{"id": "scene0409_00_0", "scene_id": "scene0409_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": "validation"}
{"id": "scene0409_00_1", "scene_id": "scene0409_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 coffee kettle in this scene?\nA. microwave\nB. dishwasher\nC. kitchen cabinets\nD. coffee maker", "choices": ["A. microwave", "B. dishwasher", "C. kitchen cabinets", "D. coffee maker"], "answer": "A. microwave", "target": "coffee kettle", "split": "validation"}
{"id": "scene0409_00_2", "scene_id": "scene0409_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 coffee kettle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The coffee kettle is next to the microwave, about 7 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 above the dishwasher, about 39 cm away.", "target": "coffee kettle", "split": "validation"}
{"id": "scene0409_00_3", "scene_id": "scene0409_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 coffee kettle in structured JSON format.", "choices": null, "answer": "{\"target\": \"coffee kettle\", \"support_surface\": \"coffee maker\", \"references\": [{\"object\": \"microwave\", \"relation\": \"next_to\", \"distance_cm\": 6.7}, {\"object\": \"kitchen cabinets\", \"relation\": \"side_above\", \"distance_cm\": 25.7}]}", "target": "coffee kettle", "split": "validation"}
{"id": "scene0409_00_4", "scene_id": "scene0409_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 fire extinguisher appear?", "choices": null, "answer": "Yes", "target": "fire extinguisher", "split": "validation"}
{"id": "scene0409_00_5", "scene_id": "scene0409_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. dishwasher\nC. coffee maker\nD. coffee kettle", "choices": ["A. chair", "B. dishwasher", "C. coffee maker", "D. coffee kettle"], "answer": "A. chair", "target": "fire extinguisher", "split": "validation"}
{"id": "scene0409_00_6", "scene_id": "scene0409_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 chair, about 55 cm away.", "target": "fire extinguisher", "split": "validation"}
{"id": "scene0409_00_7", "scene_id": "scene0409_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_below\", \"distance_cm\": 54.6}]}", "target": "fire extinguisher", "split": "validation"}
{"id": "scene0409_00_8", "scene_id": "scene0409_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 stuffed animal appear?", "choices": null, "answer": "No", "target": "stuffed animal", "split": "validation"}
{"id": "scene0409_00_9", "scene_id": "scene0409_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. coffee maker\nB. soap dispenser\nC. dishwasher\nD. kitchen cabinet", "choices": ["A. coffee maker", "B. soap dispenser", "C. dishwasher", "D. kitchen cabinet"], "answer": "A. coffee maker", "target": "plate", "split": "validation"}
{"id": "scene0409_00_10", "scene_id": "scene0409_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 below the coffee maker, about 23 cm away. It is also to the side of and above the kitchen cabinet, about 43 cm away. It is further to the side of and above the dishwasher, about 43 cm away.", "target": "plate", "split": "validation"}
{"id": "scene0409_00_11", "scene_id": "scene0409_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\": \"kitchen counter\", \"references\": [{\"object\": \"coffee maker\", \"relation\": \"side_below\", \"distance_cm\": 22.8}, {\"object\": \"kitchen cabinet\", \"relation\": \"side_above\", \"distance_cm\": 42.6}]}", "target": "plate", "split": "validation"}
{"id": "scene0409_00_12", "scene_id": "scene0409_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 soap dispenser appear?", "choices": null, "answer": "Yes", "target": "soap dispenser", "split": "validation"}
{"id": "scene0409_00_13", "scene_id": "scene0409_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 dispenser in this scene?\nA. chair\nB. kitchen counter\nC. plate\nD. kitchen cabinet", "choices": ["A. chair", "B. kitchen counter", "C. plate", "D. kitchen cabinet"], "answer": "B. kitchen counter", "target": "soap dispenser", "split": "validation"}
{"id": "scene0409_00_14", "scene_id": "scene0409_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 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 13 cm away. It is also to the side of and above the plate, about 61 cm away. It is further to the side of and above the kitchen cabinet, about 66 cm away.", "target": "soap dispenser", "split": "validation"}
{"id": "scene0409_00_15", "scene_id": "scene0409_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 dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dispenser\", \"support_surface\": \"kitchen counter\", \"references\": [{\"object\": \"kitchen counter\", \"relation\": \"side_above\", \"distance_cm\": 13.4}, {\"object\": \"plate\", \"relation\": \"side_above\", \"distance_cm\": 60.7}]}", "target": "soap dispenser", "split": "validation"}
{"id": "scene0430_00_0", "scene_id": "scene0430_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": "validation"}
{"id": "scene0434_00_0", "scene_id": "scene0434_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": "validation"}
{"id": "scene0434_00_1", "scene_id": "scene0434_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. ledge\nB. bathroom vanity\nC. towel\nD. bathtub", "choices": ["A. ledge", "B. bathroom vanity", "C. towel", "D. bathtub"], "answer": "D. bathtub", "target": "soap dish", "split": "validation"}
{"id": "scene0434_00_2", "scene_id": "scene0434_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 26 cm away. It is also to the side of and above the towel, about 62 cm away.", "target": "soap dish", "split": "validation"}
{"id": "scene0434_00_3", "scene_id": "scene0434_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\": 26.2}, {\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 61.5}]}", "target": "soap dish", "split": "validation"}
{"id": "scene0438_00_0", "scene_id": "scene0438_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": "validation"}
{"id": "scene0438_00_1", "scene_id": "scene0438_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. nightstand\nB. tissue box\nC. dresser\nD. ball", "choices": ["A. nightstand", "B. tissue box", "C. dresser", "D. ball"], "answer": "D. ball", "target": "bag", "split": "validation"}
{"id": "scene0438_00_2", "scene_id": "scene0438_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 ball. It is also to the side of and below the dresser, about 35 cm away. It is further to the side of and below the nightstand, about 39 cm away.", "target": "bag", "split": "validation"}
{"id": "scene0438_00_3", "scene_id": "scene0438_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\": \"ball\", \"relation\": \"next_to\", \"distance_cm\": 2.4}, {\"object\": \"dresser\", \"relation\": \"side_below\", \"distance_cm\": 35.3}]}", "target": "bag", "split": "validation"}
{"id": "scene0446_00_0", "scene_id": "scene0446_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": "validation"}
{"id": "scene0446_00_1", "scene_id": "scene0446_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. radiator\nB. rod\nC. towel\nD. mat", "choices": ["A. radiator", "B. rod", "C. towel", "D. mat"], "answer": "C. towel", "target": "toilet paper", "split": "validation"}
{"id": "scene0446_00_2", "scene_id": "scene0446_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 towel, about 25 cm away. It is also to the side of and above the radiator, about 29 cm away. It is further to the side of and above the mat, about 44 cm away.", "target": "toilet paper", "split": "validation"}
{"id": "scene0446_00_3", "scene_id": "scene0446_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\": \"radiator\", \"references\": [{\"object\": \"towel\", \"relation\": \"side_below\", \"distance_cm\": 24.8}, {\"object\": \"radiator\", \"relation\": \"side_above\", \"distance_cm\": 29.3}]}", "target": "toilet paper", "split": "validation"}
{"id": "scene0476_00_0", "scene_id": "scene0476_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": "validation"}
{"id": "scene0476_00_1", "scene_id": "scene0476_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. office chair\nC. keyboard\nD. file cabinet", "choices": ["A. monitor", "B. office chair", "C. keyboard", "D. file cabinet"], "answer": "D. file cabinet", "target": "backpack", "split": "validation"}
{"id": "scene0476_00_2", "scene_id": "scene0476_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 file cabinet, about 17 cm away. It is also to the side of and below the monitor, about 39 cm away. It is further to the side of and above the office chair, about 44 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0476_00_3", "scene_id": "scene0476_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\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 17.1}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 39.2}]}", "target": "backpack", "split": "validation"}
{"id": "scene0476_00_4", "scene_id": "scene0476_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": "validation"}
{"id": "scene0476_00_5", "scene_id": "scene0476_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. monitor\nB. clothing\nC. file cabinet\nD. office chair", "choices": ["A. monitor", "B. clothing", "C. file cabinet", "D. office chair"], "answer": "C. file cabinet", "target": "book", "split": "validation"}
{"id": "scene0476_00_6", "scene_id": "scene0476_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 file cabinet, about 25 cm away. It is also to the side of and above the office chair, about 27 cm away. It is further to the side of and above the clothing, about 30 cm away.", "target": "book", "split": "validation"}
{"id": "scene0476_00_7", "scene_id": "scene0476_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\": \"file cabinet\", \"relation\": \"side_above\", \"distance_cm\": 24.6}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 27.0}]}", "target": "book", "split": "validation"}
{"id": "scene0476_00_8", "scene_id": "scene0476_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": "Yes", "target": "cup", "split": "validation"}
{"id": "scene0476_00_9", "scene_id": "scene0476_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 cup in this scene?\nA. keyboard\nB. telephone\nC. monitor\nD. office chair", "choices": ["A. keyboard", "B. telephone", "C. monitor", "D. office chair"], "answer": "A. keyboard", "target": "cup", "split": "validation"}
{"id": "scene0476_00_10", "scene_id": "scene0476_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 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 10 cm away. It is further near the telephone, about 31 cm away.", "target": "cup", "split": "validation"}
{"id": "scene0476_00_11", "scene_id": "scene0476_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 cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 9.6}, {\"object\": \"telephone\", \"relation\": \"near\", \"distance_cm\": 30.5}]}", "target": "cup", "split": "validation"}
{"id": "scene0477_00_0", "scene_id": "scene0477_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": "validation"}
{"id": "scene0477_00_1", "scene_id": "scene0477_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. dresser\nB. shoes\nC. clothing\nD. lamp", "choices": ["A. dresser", "B. shoes", "C. clothing", "D. lamp"], "answer": "B. shoes", "target": "backpack", "split": "validation"}
{"id": "scene0477_00_2", "scene_id": "scene0477_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 clothing, about 36 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0477_00_3", "scene_id": "scene0477_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\": \"shoes\", \"relation\": \"side_above\", \"distance_cm\": 4.8}, {\"object\": \"clothing\", \"relation\": \"side_below\", \"distance_cm\": 36.4}]}", "target": "backpack", "split": "validation"}
{"id": "scene0477_00_4", "scene_id": "scene0477_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": "validation"}
{"id": "scene0477_00_5", "scene_id": "scene0477_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. backpack\nB. pillow\nC. lamp\nD. clothing", "choices": ["A. backpack", "B. pillow", "C. lamp", "D. clothing"], "answer": "B. pillow", "target": "telephone", "split": "validation"}
{"id": "scene0477_00_6", "scene_id": "scene0477_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 pillow, about 22 cm away. It is further to the side of and below the lamp, about 63 cm away.", "target": "telephone", "split": "validation"}
{"id": "scene0477_00_7", "scene_id": "scene0477_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\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 21.9}, {\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 63.4}]}", "target": "telephone", "split": "validation"}
{"id": "scene0477_00_8", "scene_id": "scene0477_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": "validation"}
{"id": "scene0477_00_9", "scene_id": "scene0477_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. picture\nB. towel\nC. backpack\nD. trash can", "choices": ["A. picture", "B. towel", "C. backpack", "D. trash can"], "answer": "B. towel", "target": "toilet paper", "split": "validation"}
{"id": "scene0477_00_10", "scene_id": "scene0477_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 below the towel, about 19 cm away. It is also to the side of and below the picture, about 36 cm away. It is further to the side of and above the trash can, about 57 cm away.", "target": "toilet paper", "split": "validation"}
{"id": "scene0477_00_11", "scene_id": "scene0477_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\": \"towel\", \"relation\": \"side_below\", \"distance_cm\": 19.2}, {\"object\": \"picture\", \"relation\": \"side_below\", \"distance_cm\": 36.0}]}", "target": "toilet paper", "split": "validation"}
{"id": "scene0477_00_12", "scene_id": "scene0477_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 hair dryer appear?", "choices": null, "answer": "No", "target": "hair dryer", "split": "validation"}
{"id": "scene0477_00_13", "scene_id": "scene0477_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 tray in structured JSON format.", "choices": null, "answer": "{\"target\": \"tray\", \"support_surface\": \"dresser\", \"references\": []}", "target": "tray", "split": "validation"}
{"id": "scene0508_00_0", "scene_id": "scene0508_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": "validation"}
{"id": "scene0508_00_1", "scene_id": "scene0508_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. telephone\nB. chair\nC. thermostat\nD. trash can", "choices": ["A. telephone", "B. chair", "C. thermostat", "D. trash can"], "answer": "D. trash can", "target": "backpack", "split": "validation"}
{"id": "scene0508_00_2", "scene_id": "scene0508_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 39 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0508_00_3", "scene_id": "scene0508_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\": \"trash can\", \"references\": [{\"object\": \"trash can\", \"relation\": \"side_below\", \"distance_cm\": 38.8}]}", "target": "backpack", "split": "validation"}
{"id": "scene0508_00_4", "scene_id": "scene0508_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": "validation"}
{"id": "scene0508_00_5", "scene_id": "scene0508_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. thermostat\nB. trash can\nC. backpack\nD. chair", "choices": ["A. thermostat", "B. trash can", "C. backpack", "D. chair"], "answer": "A. thermostat", "target": "telephone", "split": "validation"}
{"id": "scene0508_00_6", "scene_id": "scene0508_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 below the thermostat, about 16 cm away. It is also to the side of and above the chair, about 50 cm away.", "target": "telephone", "split": "validation"}
{"id": "scene0508_00_7", "scene_id": "scene0508_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\": \"chair\", \"references\": [{\"object\": \"thermostat\", \"relation\": \"side_below\", \"distance_cm\": 15.6}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 50.2}]}", "target": "telephone", "split": "validation"}
{"id": "scene0509_00_0", "scene_id": "scene0509_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": "validation"}
{"id": "scene0509_00_1", "scene_id": "scene0509_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. bathroom vanity\nB. sink\nC. mirror\nD. hair dryer", "choices": ["A. bathroom vanity", "B. sink", "C. mirror", "D. hair dryer"], "answer": "C. mirror", "target": "bag", "split": "validation"}
{"id": "scene0509_00_2", "scene_id": "scene0509_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 mirror. It is also to the side of and above the sink, about 6 cm away.", "target": "bag", "split": "validation"}
{"id": "scene0509_00_3", "scene_id": "scene0509_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\": \"mirror\", \"relation\": \"under\", \"distance_cm\": 2.5}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 6.0}]}", "target": "bag", "split": "validation"}
{"id": "scene0509_00_4", "scene_id": "scene0509_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": "validation"}
{"id": "scene0509_00_5", "scene_id": "scene0509_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. towel\nB. toilet paper\nC. tissue box\nD. sink", "choices": ["A. towel", "B. toilet paper", "C. tissue box", "D. sink"], "answer": "A. towel", "target": "hair dryer", "split": "validation"}
{"id": "scene0509_00_6", "scene_id": "scene0509_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 towel. It is also to the side of and above the toilet paper, about 27 cm away. It is further to the side of and below the sink, about 34 cm away.", "target": "hair dryer", "split": "validation"}
{"id": "scene0509_00_7", "scene_id": "scene0509_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\": \"toilet paper\", \"references\": [{\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 2.5}, {\"object\": \"toilet paper\", \"relation\": \"side_above\", \"distance_cm\": 27.5}]}", "target": "hair dryer", "split": "validation"}
{"id": "scene0509_00_8", "scene_id": "scene0509_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": "validation"}
{"id": "scene0509_00_9", "scene_id": "scene0509_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. toilet paper\nB. sink\nC. hair dryer\nD. towel", "choices": ["A. toilet paper", "B. sink", "C. hair dryer", "D. towel"], "answer": "B. sink", "target": "tissue box", "split": "validation"}
{"id": "scene0509_00_10", "scene_id": "scene0509_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 sink, about 16 cm away. It is also to the side of and above the towel, about 30 cm away. It is further to the side of and above the hair dryer, about 34 cm away.", "target": "tissue box", "split": "validation"}
{"id": "scene0509_00_11", "scene_id": "scene0509_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\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 15.5}, {\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 30.1}]}", "target": "tissue box", "split": "validation"}
{"id": "scene0509_00_12", "scene_id": "scene0509_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": "Yes", "target": "toilet paper", "split": "validation"}
{"id": "scene0509_00_13", "scene_id": "scene0509_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. toilet\nB. towel\nC. bathroom vanity\nD. trash can", "choices": ["A. toilet", "B. towel", "C. bathroom vanity", "D. trash can"], "answer": "D. trash can", "target": "toilet paper", "split": "validation"}
{"id": "scene0509_00_14", "scene_id": "scene0509_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 above the trash can, about 8 cm away. It is also near the toilet, about 16 cm away. It is further to the side of and below the bathroom vanity, about 26 cm away.", "target": "toilet paper", "split": "validation"}
{"id": "scene0509_00_15", "scene_id": "scene0509_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\": \"tissue box\", \"references\": [{\"object\": \"trash can\", \"relation\": \"above\", \"distance_cm\": 8.3}, {\"object\": \"toilet\", \"relation\": \"near\", \"distance_cm\": 15.9}]}", "target": "toilet paper", "split": "validation"}
{"id": "scene0518_00_0", "scene_id": "scene0518_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": "validation"}
{"id": "scene0518_00_1", "scene_id": "scene0518_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. picture\nB. desk\nC. monitor\nD. office chair", "choices": ["A. picture", "B. desk", "C. monitor", "D. office chair"], "answer": "B. desk", "target": "backpack", "split": "validation"}
{"id": "scene0518_00_2", "scene_id": "scene0518_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 desk, about 5 cm away. It is also to the side of and below the picture, about 53 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0518_00_3", "scene_id": "scene0518_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\": \"side_above\", \"distance_cm\": 4.6}, {\"object\": \"picture\", \"relation\": \"side_below\", \"distance_cm\": 53.0}]}", "target": "backpack", "split": "validation"}
{"id": "scene0522_00_0", "scene_id": "scene0522_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": "validation"}
{"id": "scene0522_00_1", "scene_id": "scene0522_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. clothes dryer\nB. cabinet\nC. basket\nD. bag", "choices": ["A. clothes dryer", "B. cabinet", "C. basket", "D. bag"], "answer": "D. bag", "target": "purse", "split": "validation"}
{"id": "scene0522_00_2", "scene_id": "scene0522_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 counter. It is also next to the bag, about 4 cm away. It is further to the side of and above the clothes dryer, about 8 cm away.", "target": "purse", "split": "validation"}
{"id": "scene0522_00_3", "scene_id": "scene0522_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\": \"counter\", \"references\": [{\"object\": \"bag\", \"relation\": \"next_to\", \"distance_cm\": 4.0}, {\"object\": \"clothes dryer\", \"relation\": \"side_above\", \"distance_cm\": 7.6}]}", "target": "purse", "split": "validation"}
{"id": "scene0532_00_0", "scene_id": "scene0532_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": "validation"}
{"id": "scene0532_00_1", "scene_id": "scene0532_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. monitor\nB. chair\nC. desk\nD. cable", "choices": ["A. monitor", "B. chair", "C. desk", "D. cable"], "answer": "D. cable", "target": "fire extinguisher", "split": "validation"}
{"id": "scene0532_00_2", "scene_id": "scene0532_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 cable, about 81 cm away.", "target": "fire extinguisher", "split": "validation"}
{"id": "scene0532_00_3", "scene_id": "scene0532_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\": \"cable\", \"references\": [{\"object\": \"cable\", \"relation\": \"side_below\", \"distance_cm\": 80.8}]}", "target": "fire extinguisher", "split": "validation"}
{"id": "scene0535_00_0", "scene_id": "scene0535_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": "validation"}
{"id": "scene0535_00_1", "scene_id": "scene0535_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. backpack\nB. desk\nC. monitor\nD. chair", "choices": ["A. backpack", "B. desk", "C. monitor", "D. chair"], "answer": "D. chair", "target": "book", "split": "validation"}
{"id": "scene0535_00_2", "scene_id": "scene0535_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 above the chair, about 38 cm away.", "target": "book", "split": "validation"}
{"id": "scene0535_00_3", "scene_id": "scene0535_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\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 38.0}]}", "target": "book", "split": "validation"}
{"id": "scene0538_00_0", "scene_id": "scene0538_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": "validation"}
{"id": "scene0538_00_1", "scene_id": "scene0538_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. bathroom vanity\nC. mirror\nD. ledge", "choices": ["A. toilet", "B. bathroom vanity", "C. mirror", "D. ledge"], "answer": "A. toilet", "target": "plunger", "split": "validation"}
{"id": "scene0538_00_2", "scene_id": "scene0538_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 near the toilet, about 20 cm away.", "target": "plunger", "split": "validation"}
{"id": "scene0538_00_3", "scene_id": "scene0538_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\": \"near\", \"distance_cm\": 19.8}]}", "target": "plunger", "split": "validation"}
{"id": "scene0538_00_4", "scene_id": "scene0538_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": "validation"}
{"id": "scene0538_00_5", "scene_id": "scene0538_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. paper towel dispenser\nB. sink\nC. ledge\nD. bathroom vanity", "choices": ["A. paper towel dispenser", "B. sink", "C. ledge", "D. bathroom vanity"], "answer": "C. ledge", "target": "soap dispenser", "split": "validation"}
{"id": "scene0538_00_6", "scene_id": "scene0538_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 ledge, about 14 cm away. It is also to the side of and below the paper towel dispenser, about 22 cm away. It is further to the side of and above the bathroom vanity, about 30 cm away.", "target": "soap dispenser", "split": "validation"}
{"id": "scene0538_00_7", "scene_id": "scene0538_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\": \"paper towel dispenser\", \"references\": [{\"object\": \"ledge\", \"relation\": \"side_above\", \"distance_cm\": 13.9}, {\"object\": \"paper towel dispenser\", \"relation\": \"side_below\", \"distance_cm\": 21.8}]}", "target": "soap dispenser", "split": "validation"}
{"id": "scene0547_00_0", "scene_id": "scene0547_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": "validation"}
{"id": "scene0547_00_1", "scene_id": "scene0547_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. dresser\nB. laundry hamper\nC. bag\nD. computer tower", "choices": ["A. dresser", "B. laundry hamper", "C. bag", "D. computer tower"], "answer": "A. dresser", "target": "backpack", "split": "validation"}
{"id": "scene0547_00_2", "scene_id": "scene0547_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 dresser, about 16 cm away. It is also to the side of and above the laundry hamper, about 24 cm away. It is further to the side of and below the computer tower, about 34 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0547_00_3", "scene_id": "scene0547_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\": \"dresser\", \"relation\": \"near\", \"distance_cm\": 15.8}, {\"object\": \"laundry hamper\", \"relation\": \"side_above\", \"distance_cm\": 23.9}]}", "target": "backpack", "split": "validation"}
{"id": "scene0547_00_4", "scene_id": "scene0547_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": "validation"}
{"id": "scene0547_00_5", "scene_id": "scene0547_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. clothing\nB. backpack\nC. clothes\nD. chair", "choices": ["A. clothing", "B. backpack", "C. clothes", "D. chair"], "answer": "A. clothing", "target": "bag", "split": "validation"}
{"id": "scene0547_00_6", "scene_id": "scene0547_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 clothing, about 32 cm away. It is also near the backpack, about 40 cm away. It is further to the side of and below the clothes, about 57 cm away.", "target": "bag", "split": "validation"}
{"id": "scene0547_00_7", "scene_id": "scene0547_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\": \"chair\", \"references\": [{\"object\": \"clothing\", \"relation\": \"side_below\", \"distance_cm\": 32.1}, {\"object\": \"backpack\", \"relation\": \"near\", \"distance_cm\": 40.4}]}", "target": "bag", "split": "validation"}
{"id": "scene0547_00_8", "scene_id": "scene0547_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": "No", "target": "plate", "split": "validation"}
{"id": "scene0547_00_9", "scene_id": "scene0547_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. chair\nB. shelf\nC. mini fridge\nD. monitor", "choices": ["A. chair", "B. shelf", "C. mini fridge", "D. monitor"], "answer": "B. shelf", "target": "bowl", "split": "validation"}
{"id": "scene0547_00_10", "scene_id": "scene0547_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 desk. It is also under the shelf, about 3 cm away. It is further to the side of and below the monitor, about 18 cm away.", "target": "bowl", "split": "validation"}
{"id": "scene0547_00_11", "scene_id": "scene0547_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\": \"desk\", \"references\": [{\"object\": \"shelf\", \"relation\": \"under\", \"distance_cm\": 2.5}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 18.1}]}", "target": "bowl", "split": "validation"}
{"id": "scene0547_00_12", "scene_id": "scene0547_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 water pitcher appear?", "choices": null, "answer": "Yes", "target": "water pitcher", "split": "validation"}
{"id": "scene0547_00_13", "scene_id": "scene0547_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 water pitcher in this scene?\nA. shelf\nB. bowl\nC. chair\nD. pillow", "choices": ["A. shelf", "B. bowl", "C. chair", "D. pillow"], "answer": "A. shelf", "target": "water pitcher", "split": "validation"}
{"id": "scene0547_00_14", "scene_id": "scene0547_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 water pitcher in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water pitcher is to the side of and below the shelf, about 32 cm away. It is also to the side of and above the bowl, about 45 cm away. It is further to the side of and above the pillow, about 45 cm away.", "target": "water pitcher", "split": "validation"}
{"id": "scene0547_00_15", "scene_id": "scene0547_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 water pitcher in structured JSON format.", "choices": null, "answer": "{\"target\": \"water pitcher\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 32.0}, {\"object\": \"bowl\", \"relation\": \"side_above\", \"distance_cm\": 45.1}]}", "target": "water pitcher", "split": "validation"}
{"id": "scene0563_00_0", "scene_id": "scene0563_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": "validation"}
{"id": "scene0563_00_1", "scene_id": "scene0563_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. desk\nD. chair", "choices": ["A. keyboard", "B. monitor", "C. desk", "D. chair"], "answer": "D. chair", "target": "backpack", "split": "validation"}
{"id": "scene0564_00_0", "scene_id": "scene0564_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": "validation"}
{"id": "scene0564_00_1", "scene_id": "scene0564_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. sign\nB. sink\nC. trash can\nD. bathroom vanity", "choices": ["A. sign", "B. sink", "C. trash can", "D. bathroom vanity"], "answer": "D. bathroom vanity", "target": "broom", "split": "validation"}
{"id": "scene0564_00_2", "scene_id": "scene0564_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 bathroom vanity, about 15 cm away. It is also to the side of and below the sink, about 38 cm away. It is further to the side of and above the trash can, about 54 cm away.", "target": "broom", "split": "validation"}
{"id": "scene0564_00_3", "scene_id": "scene0564_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\": \"sign\", \"references\": [{\"object\": \"bathroom vanity\", \"relation\": \"side_below\", \"distance_cm\": 15.1}, {\"object\": \"sink\", \"relation\": \"side_below\", \"distance_cm\": 38.0}]}", "target": "broom", "split": "validation"}
{"id": "scene0566_00_0", "scene_id": "scene0566_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": "validation"}
{"id": "scene0566_00_1", "scene_id": "scene0566_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. table\nD. desk", "choices": ["A. chair", "B. monitor", "C. table", "D. desk"], "answer": "C. table", "target": "backpack", "split": "validation"}
{"id": "scene0566_00_2", "scene_id": "scene0566_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 table, about 17 cm away.", "target": "backpack", "split": "validation"}
{"id": "scene0566_00_3", "scene_id": "scene0566_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\": \"near\", \"distance_cm\": 16.6}]}", "target": "backpack", "split": "validation"}
{"id": "scene0571_00_0", "scene_id": "scene0571_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": "validation"}
{"id": "scene0571_00_1", "scene_id": "scene0571_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": "No", "target": "book", "split": "validation"}
{"id": "scene0571_00_2", "scene_id": "scene0571_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 plunger in this scene?\nA. toilet\nB. shampoo bottle\nC. backpack\nD. bathtub", "choices": ["A. toilet", "B. shampoo bottle", "C. backpack", "D. bathtub"], "answer": "A. toilet", "target": "plunger", "split": "validation"}
{"id": "scene0571_00_3", "scene_id": "scene0571_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 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 14 cm away. It is also to the side of and below the shampoo bottle, about 69 cm away.", "target": "plunger", "split": "validation"}
{"id": "scene0571_00_4", "scene_id": "scene0571_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 plunger in structured JSON format.", "choices": null, "answer": "{\"target\": \"plunger\", \"support_surface\": \"bathtub\", \"references\": [{\"object\": \"toilet\", \"relation\": \"side_below\", \"distance_cm\": 13.7}, {\"object\": \"shampoo bottle\", \"relation\": \"side_below\", \"distance_cm\": 69.2}]}", "target": "plunger", "split": "validation"}
{"id": "scene0571_00_5", "scene_id": "scene0571_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 laptop appear?", "choices": null, "answer": "No", "target": "laptop", "split": "validation"}
{"id": "scene0571_00_6", "scene_id": "scene0571_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 soap dish in this scene?\nA. bathtub\nB. backpack\nC. shampoo bottle\nD. conditioner bottle", "choices": ["A. bathtub", "B. backpack", "C. shampoo bottle", "D. conditioner bottle"], "answer": "A. bathtub", "target": "soap dish", "split": "validation"}
{"id": "scene0571_00_7", "scene_id": "scene0571_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 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 above the conditioner bottle, about 49 cm away. It is further to the side of and above the shampoo bottle, about 54 cm away.", "target": "soap dish", "split": "validation"}
{"id": "scene0571_00_8", "scene_id": "scene0571_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 soap dish in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap dish\", \"support_surface\": \"shower walls\", \"references\": [{\"object\": \"bathtub\", \"relation\": \"side_above\", \"distance_cm\": 20.3}, {\"object\": \"conditioner bottle\", \"relation\": \"side_above\", \"distance_cm\": 49.3}]}", "target": "soap dish", "split": "validation"}
{"id": "scene0580_00_0", "scene_id": "scene0580_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": "validation"}
{"id": "scene0580_00_1", "scene_id": "scene0580_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 dish appear?", "choices": null, "answer": "No", "target": "soap dish", "split": "validation"}
{"id": "scene0580_00_2", "scene_id": "scene0580_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 hat in this scene?\nA. backpack\nB. bag\nC. bookshelf\nD. trash can", "choices": ["A. backpack", "B. bag", "C. bookshelf", "D. trash can"], "answer": "C. bookshelf", "target": "hat", "split": "validation"}
{"id": "scene0580_00_3", "scene_id": "scene0580_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 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 bookshelf, about 16 cm away. It is further to the side of and above the trash can, about 59 cm away.", "target": "hat", "split": "validation"}
{"id": "scene0580_00_4", "scene_id": "scene0580_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 hat in structured JSON format.", "choices": null, "answer": "{\"target\": \"hat\", \"support_surface\": \"dresser\", \"references\": [{\"object\": \"bookshelf\", \"relation\": \"side_below\", \"distance_cm\": 16.4}, {\"object\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 59.2}]}", "target": "hat", "split": "validation"}
{"id": "scene0582_00_0", "scene_id": "scene0582_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": "validation"}
{"id": "scene0582_00_1", "scene_id": "scene0582_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. telephone\nB. office chair\nC. monitor\nD. computer tower", "choices": ["A. telephone", "B. office chair", "C. monitor", "D. computer tower"], "answer": "B. office chair", "target": "keyboard", "split": "validation"}
{"id": "scene0582_00_2", "scene_id": "scene0582_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 office chair, about 12 cm away. It is further to the side of and below the monitor, about 18 cm away.", "target": "keyboard", "split": "validation"}
{"id": "scene0582_00_3", "scene_id": "scene0582_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\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 12.3}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 18.1}]}", "target": "keyboard", "split": "validation"}
{"id": "scene0582_00_4", "scene_id": "scene0582_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": "validation"}
{"id": "scene0582_00_5", "scene_id": "scene0582_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. keyboard\nB. telephone\nC. monitor\nD. book", "choices": ["A. keyboard", "B. telephone", "C. monitor", "D. book"], "answer": "A. keyboard", "target": "mouse", "split": "validation"}
{"id": "scene0582_00_6", "scene_id": "scene0582_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 desk. It is also next to the keyboard. It is further to the side of and below the monitor, about 11 cm away.", "target": "mouse", "split": "validation"}
{"id": "scene0582_00_7", "scene_id": "scene0582_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\": \"desk\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 2.2}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 10.7}]}", "target": "mouse", "split": "validation"}
{"id": "scene0582_00_8", "scene_id": "scene0582_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": "Yes", "target": "power strip", "split": "validation"}
{"id": "scene0582_00_9", "scene_id": "scene0582_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 power strip in this scene?\nA. box\nB. book\nC. cabinet\nD. desk", "choices": ["A. box", "B. book", "C. cabinet", "D. desk"], "answer": "D. desk", "target": "power strip", "split": "validation"}
{"id": "scene0582_00_10", "scene_id": "scene0582_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 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 desk, about 16 cm away. It is also to the side of and below the cabinet, about 53 cm away.", "target": "power strip", "split": "validation"}
{"id": "scene0582_00_11", "scene_id": "scene0582_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 power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 15.9}, {\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 52.7}]}", "target": "power strip", "split": "validation"}
{"id": "00dd871005_0", "scene_id": "00dd871005", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "00dd871005_1", "scene_id": "00dd871005", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. pasta box\nB. rolling cabinet\nC. polaroid camera\nD. cardboard box", "choices": ["A. pasta box", "B. rolling cabinet", "C. polaroid camera", "D. cardboard box"], "answer": "D. cardboard box", "target": "hairbrush", "split": "validation"}
{"id": "00dd871005_2", "scene_id": "00dd871005", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 to the side of and below the cardboard box, about 7 cm away. It is also to the side of and above the rolling cabinet, about 17 cm away. It is further to the side of and above the polaroid camera, about 21 cm away.", "target": "hairbrush", "split": "validation"}
{"id": "00dd871005_3", "scene_id": "00dd871005", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"cardboard box\", \"relation\": \"side_below\", \"distance_cm\": 6.6}, {\"object\": \"rolling cabinet\", \"relation\": \"side_above\", \"distance_cm\": 16.7}]}", "target": "hairbrush", "split": "validation"}
{"id": "00dd871005_4", "scene_id": "00dd871005", "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": "validation"}
{"id": "00dd871005_5", "scene_id": "00dd871005", "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. pillow\nC. shelf\nD. projector", "choices": ["A. light switch", "B. pillow", "C. shelf", "D. projector"], "answer": "C. shelf", "target": "phone charger", "split": "validation"}
{"id": "00dd871005_6", "scene_id": "00dd871005", "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 near the shelf, about 28 cm away. It is also to the side of and below the projector, about 33 cm away. It is further above the light switch, about 50 cm away.", "target": "phone charger", "split": "validation"}
{"id": "00dd871005_7", "scene_id": "00dd871005", "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\": \"shelf\", \"relation\": \"near\", \"distance_cm\": 27.8}, {\"object\": \"projector\", \"relation\": \"side_below\", \"distance_cm\": 33.5}]}", "target": "phone charger", "split": "validation"}
{"id": "2e74812d00_0", "scene_id": "2e74812d00", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "2e74812d00_1", "scene_id": "2e74812d00", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. ball\nB. split\nC. computer tower\nD. power socket", "choices": ["A. ball", "B. split", "C. computer tower", "D. power socket"], "answer": "D. power socket", "target": "cable", "split": "validation"}
{"id": "2e74812d00_2", "scene_id": "2e74812d00", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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. It is also to the side of and above the computer tower, about 20 cm away. It is further to the side of and above the ball, about 22 cm away.", "target": "cable", "split": "validation"}
{"id": "2e74812d00_3", "scene_id": "2e74812d00", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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 chair\", \"references\": [{\"object\": \"power socket\", \"relation\": \"under\", \"distance_cm\": 2.1}, {\"object\": \"computer tower\", \"relation\": \"side_above\", \"distance_cm\": 20.4}]}", "target": "cable", "split": "validation"}
{"id": "2e74812d00_4", "scene_id": "2e74812d00", "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": "validation"}
{"id": "2e74812d00_5", "scene_id": "2e74812d00", "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\nB. plant pot\nC. plant\nD. bottle", "choices": ["A. pen", "B. plant pot", "C. plant", "D. bottle"], "answer": "A. pen", "target": "cup", "split": "validation"}
{"id": "2e74812d00_6", "scene_id": "2e74812d00", "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 pen, about 3 cm away. It is further next to the bottle, about 6 cm away.", "target": "cup", "split": "validation"}
{"id": "2e74812d00_7", "scene_id": "2e74812d00", "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\": \"pen\", \"relation\": \"next_to\", \"distance_cm\": 2.6}, {\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 5.7}]}", "target": "cup", "split": "validation"}
{"id": "2e74812d00_8", "scene_id": "2e74812d00", "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 extension appear?", "choices": null, "answer": "No", "target": "power extension", "split": "validation"}
{"id": "2e74812d00_9", "scene_id": "2e74812d00", "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. office chair\nB. food container\nC. mug\nD. monitor", "choices": ["A. office chair", "B. food container", "C. mug", "D. monitor"], "answer": "D. monitor", "target": "glasses case", "split": "validation"}
{"id": "2e74812d00_10", "scene_id": "2e74812d00", "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 under the monitor, about 4 cm away. It is further next to the mug, about 10 cm away.", "target": "glasses case", "split": "validation"}
{"id": "2e74812d00_11", "scene_id": "2e74812d00", "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\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 3.8}, {\"object\": \"mug\", \"relation\": \"next_to\", \"distance_cm\": 10.0}]}", "target": "glasses case", "split": "validation"}
{"id": "2e74812d00_12", "scene_id": "2e74812d00", "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 eraser appear?", "choices": null, "answer": "No", "target": "eraser", "split": "validation"}
{"id": "2e74812d00_13", "scene_id": "2e74812d00", "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 headphone case in this scene?\nA. table\nB. monitor\nC. keyboard\nD. book", "choices": ["A. table", "B. monitor", "C. keyboard", "D. book"], "answer": "A. table", "target": "headphone case", "split": "validation"}
{"id": "2e74812d00_14", "scene_id": "2e74812d00", "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 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 keyboard, about 11 cm away. It is further near the book, about 18 cm away.", "target": "headphone case", "split": "validation"}
{"id": "2e74812d00_15", "scene_id": "2e74812d00", "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 headphone case in structured JSON format.", "choices": null, "answer": "{\"target\": \"headphone case\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 11.4}, {\"object\": \"book\", \"relation\": \"near\", \"distance_cm\": 18.4}]}", "target": "headphone case", "split": "validation"}
{"id": "2e74812d00_16", "scene_id": "2e74812d00", "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 plug appear?", "choices": null, "answer": "No", "target": "power plug", "split": "validation"}
{"id": "2e74812d00_17", "scene_id": "2e74812d00", "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. book\nB. headphone\nC. jug\nD. monitor", "choices": ["A. book", "B. headphone", "C. jug", "D. monitor"], "answer": "C. jug", "target": "mug", "split": "validation"}
{"id": "2e74812d00_18", "scene_id": "2e74812d00", "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 under the jug, about 4 cm away. It is further near the book, about 39 cm away.", "target": "mug", "split": "validation"}
{"id": "2e74812d00_19", "scene_id": "2e74812d00", "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\": \"jug\", \"relation\": \"under\", \"distance_cm\": 3.8}, {\"object\": \"book\", \"relation\": \"near\", \"distance_cm\": 39.0}]}", "target": "mug", "split": "validation"}
{"id": "2e74812d00_20", "scene_id": "2e74812d00", "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": "Yes", "target": "pen", "split": "validation"}
{"id": "2e74812d00_21", "scene_id": "2e74812d00", "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 in this scene?\nA. plant\nB. bottle\nC. plant pot\nD. cup", "choices": ["A. plant", "B. bottle", "C. plant pot", "D. cup"], "answer": "D. cup", "target": "pen", "split": "validation"}
{"id": "2e74812d00_22", "scene_id": "2e74812d00", "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 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 cup, about 3 cm away. It is further under the plant pot, about 5 cm away.", "target": "pen", "split": "validation"}
{"id": "2e74812d00_23", "scene_id": "2e74812d00", "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 in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"table\", \"references\": [{\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 2.6}, {\"object\": \"plant pot\", \"relation\": \"under\", \"distance_cm\": 4.6}]}", "target": "pen", "split": "validation"}
{"id": "354e4a0a50_0", "scene_id": "354e4a0a50", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "354e4a0a50_1", "scene_id": "354e4a0a50", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. computer tower\nC. speaker\nD. gamecontroller", "choices": ["A. monitor", "B. computer tower", "C. speaker", "D. gamecontroller"], "answer": "D. gamecontroller", "target": "keyboard", "split": "validation"}
{"id": "354e4a0a50_2", "scene_id": "354e4a0a50", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 gamecontroller, about 4 cm away. It is further next to the speaker, about 8 cm away.", "target": "keyboard", "split": "validation"}
{"id": "354e4a0a50_3", "scene_id": "354e4a0a50", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"gamecontroller\", \"relation\": \"next_to\", \"distance_cm\": 4.0}, {\"object\": \"speaker\", \"relation\": \"next_to\", \"distance_cm\": 8.3}]}", "target": "keyboard", "split": "validation"}
{"id": "354e4a0a50_4", "scene_id": "354e4a0a50", "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": "validation"}
{"id": "354e4a0a50_5", "scene_id": "354e4a0a50", "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. computer tower\nC. monitor\nD. headphones", "choices": ["A. chair", "B. computer tower", "C. monitor", "D. headphones"], "answer": "D. headphones", "target": "mouse", "split": "validation"}
{"id": "354e4a0a50_6", "scene_id": "354e4a0a50", "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 near the headphones, about 18 cm away. It is further to the side of and below the monitor, about 20 cm away.", "target": "mouse", "split": "validation"}
{"id": "354e4a0a50_7", "scene_id": "354e4a0a50", "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\": \"headphones\", \"relation\": \"near\", \"distance_cm\": 18.3}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 20.2}]}", "target": "mouse", "split": "validation"}
{"id": "354e4a0a50_8", "scene_id": "354e4a0a50", "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": "validation"}
{"id": "354e4a0a50_9", "scene_id": "354e4a0a50", "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. fan\nB. shoe\nC. clothes\nD. plushie", "choices": ["A. fan", "B. shoe", "C. clothes", "D. plushie"], "answer": "D. plushie", "target": "water bottle", "split": "validation"}
{"id": "354e4a0a50_10", "scene_id": "354e4a0a50", "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 below the plushie, about 6 cm away. It is also to the side of and below the fan, about 46 cm away. It is further to the side of and below the pillow, about 46 cm away.", "target": "water bottle", "split": "validation"}
{"id": "354e4a0a50_11", "scene_id": "354e4a0a50", "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\": \"bed\", \"references\": [{\"object\": \"plushie\", \"relation\": \"side_below\", \"distance_cm\": 5.6}, {\"object\": \"fan\", \"relation\": \"side_below\", \"distance_cm\": 45.9}]}", "target": "water bottle", "split": "validation"}
{"id": "3583c4c70f_0", "scene_id": "3583c4c70f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "3583c4c70f_1", "scene_id": "3583c4c70f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. blanket\nC. reusable shopping bag\nD. mini fridge", "choices": ["A. office chair", "B. blanket", "C. reusable shopping bag", "D. mini fridge"], "answer": "A. office chair", "target": "bottle", "split": "validation"}
{"id": "3583c4c70f_2", "scene_id": "3583c4c70f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 carpet. It is also to the side of and below the office chair, about 6 cm away. It is further to the side of and below the mini fridge, about 11 cm away.", "target": "bottle", "split": "validation"}
{"id": "3583c4c70f_3", "scene_id": "3583c4c70f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"carpet\", \"relation\": \"side_above\", \"distance_cm\": 2.4}, {\"object\": \"office chair\", \"relation\": \"side_below\", \"distance_cm\": 6.0}]}", "target": "bottle", "split": "validation"}
{"id": "3583c4c70f_4", "scene_id": "3583c4c70f", "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": "validation"}
{"id": "3583c4c70f_5", "scene_id": "3583c4c70f", "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. paper\nB. portable light\nC. calendar\nD. curtain", "choices": ["A. paper", "B. portable light", "C. calendar", "D. curtain"], "answer": "B. portable light", "target": "extension cord", "split": "validation"}
{"id": "3583c4c70f_6", "scene_id": "3583c4c70f", "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 portable light, about 3 cm away. It is also to the side of and above the curtain, about 6 cm away. It is further to the side of and above the calendar, about 16 cm away.", "target": "extension cord", "split": "validation"}
{"id": "3583c4c70f_7", "scene_id": "3583c4c70f", "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\": \"window\", \"references\": [{\"object\": \"portable light\", \"relation\": \"side_above\", \"distance_cm\": 2.8}, {\"object\": \"curtain\", \"relation\": \"side_above\", \"distance_cm\": 6.2}]}", "target": "extension cord", "split": "validation"}
{"id": "3583c4c70f_8", "scene_id": "3583c4c70f", "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": "validation"}
{"id": "3583c4c70f_9", "scene_id": "3583c4c70f", "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 on the desk. It is also to the side of and above the trolley, about 3 cm away. It is further next to the books, about 3 cm away.", "target": "laptop charger", "split": "validation"}
{"id": "3583c4c70f_10", "scene_id": "3583c4c70f", "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\": \"trolley\", \"relation\": \"side_above\", \"distance_cm\": 3.2}, {\"object\": \"books\", \"relation\": \"next_to\", \"distance_cm\": 3.3}]}", "target": "laptop charger", "split": "validation"}
{"id": "3597f00367_0", "scene_id": "3597f00367", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "3597f00367_1", "scene_id": "3597f00367", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. shelf\nB. skirting\nC. heater\nD. table", "choices": ["A. shelf", "B. skirting", "C. heater", "D. table"], "answer": "D. table", "target": "extension cord", "split": "validation"}
{"id": "3597f00367_2", "scene_id": "3597f00367", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 9 cm away. It is also to the side of and below the shelf, about 19 cm away. It is further to the side of and below the heater, about 22 cm away.", "target": "extension cord", "split": "validation"}
{"id": "3597f00367_3", "scene_id": "3597f00367", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 8.8}, {\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 19.0}]}", "target": "extension cord", "split": "validation"}
{"id": "35f2120068_0", "scene_id": "35f2120068", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "35f2120068_1", "scene_id": "35f2120068", "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 heater, about 9 cm away. It is also to the side of and below the heat pipe, about 9 cm away. It is further to the side of and below the document shredder, about 15 cm away.", "target": "extension cord", "split": "validation"}
{"id": "35f2120068_2", "scene_id": "35f2120068", "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\": \"heater\", \"references\": [{\"object\": \"heater\", \"relation\": \"side_below\", \"distance_cm\": 9.2}, {\"object\": \"heat pipe\", \"relation\": \"side_below\", \"distance_cm\": 9.3}]}", "target": "extension cord", "split": "validation"}
{"id": "35f2120068_3", "scene_id": "35f2120068", "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 pencil case appear?", "choices": null, "answer": "No", "target": "pencil case", "split": "validation"}
{"id": "35f2120068_4", "scene_id": "35f2120068", "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 handbag in this scene?\nA. cable raceway\nB. power sockets\nC. window frame\nD. wire spool", "choices": ["A. cable raceway", "B. power sockets", "C. window frame", "D. wire spool"], "answer": "B. power sockets", "target": "handbag", "split": "validation"}
{"id": "35f2120068_5", "scene_id": "35f2120068", "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 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 power sockets, about 13 cm away. It is also to the side of and below the window frame, about 30 cm away. It is further near the wire spool, about 34 cm away.", "target": "handbag", "split": "validation"}
{"id": "35f2120068_6", "scene_id": "35f2120068", "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 handbag in structured JSON format.", "choices": null, "answer": "{\"target\": \"handbag\", \"support_surface\": \"window frame\", \"references\": [{\"object\": \"power sockets\", \"relation\": \"side_above\", \"distance_cm\": 13.3}, {\"object\": \"window frame\", \"relation\": \"side_below\", \"distance_cm\": 29.7}]}", "target": "handbag", "split": "validation"}
{"id": "35f2120068_7", "scene_id": "35f2120068", "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 plug appear?", "choices": null, "answer": "No", "target": "power plug", "split": "validation"}
{"id": "35f2120068_8", "scene_id": "35f2120068", "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 next to the electronic breadboard, about 6 cm away. It is further to the side of and below the helping hands, about 6 cm away.", "target": "keyboard", "split": "validation"}
{"id": "35f2120068_9", "scene_id": "35f2120068", "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\": \"electronic breadboard\", \"relation\": \"next_to\", \"distance_cm\": 5.6}, {\"object\": \"helping hands\", \"relation\": \"side_below\", \"distance_cm\": 6.0}]}", "target": "keyboard", "split": "validation"}
{"id": "35f2120068_10", "scene_id": "35f2120068", "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 mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "validation"}
{"id": "35f2120068_11", "scene_id": "35f2120068", "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 mug in this scene?\nA. paper\nB. box\nC. 3d printer\nD. tissue box", "choices": ["A. paper", "B. box", "C. 3d printer", "D. tissue box"], "answer": "A. paper", "target": "mug", "split": "validation"}
{"id": "35f2120068_12", "scene_id": "35f2120068", "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 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 to the side of and below the 3d printer, about 14 cm away.", "target": "mug", "split": "validation"}
{"id": "35f2120068_13", "scene_id": "35f2120068", "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 mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 4.4}, {\"object\": \"3d printer\", \"relation\": \"side_below\", \"distance_cm\": 13.5}]}", "target": "mug", "split": "validation"}
{"id": "35f2120068_14", "scene_id": "35f2120068", "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 tissue box appear?", "choices": null, "answer": "Yes", "target": "tissue box", "split": "validation"}
{"id": "35f2120068_15", "scene_id": "35f2120068", "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 tissue box in this scene?\nA. mug\nB. box\nC. paper\nD. thermos", "choices": ["A. mug", "B. box", "C. paper", "D. thermos"], "answer": "B. box", "target": "tissue box", "split": "validation"}
{"id": "35f2120068_16", "scene_id": "35f2120068", "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 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 paper, about 15 cm away. It is further near the mug, about 17 cm away.", "target": "tissue box", "split": "validation"}
{"id": "35f2120068_17", "scene_id": "35f2120068", "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 tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"table\", \"references\": [{\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 15.0}, {\"object\": \"mug\", \"relation\": \"near\", \"distance_cm\": 16.8}]}", "target": "tissue box", "split": "validation"}
{"id": "3799bd47b3_0", "scene_id": "3799bd47b3", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "3799bd47b3_1", "scene_id": "3799bd47b3", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. blanket\nC. bag\nD. luggage", "choices": ["A. doorframe", "B. blanket", "C. bag", "D. luggage"], "answer": "D. luggage", "target": "bottle", "split": "validation"}
{"id": "3799bd47b3_2", "scene_id": "3799bd47b3", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 luggage, about 18 cm away. It is further to the side of and above the bag, about 37 cm away.", "target": "bottle", "split": "validation"}
{"id": "3799bd47b3_3", "scene_id": "3799bd47b3", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"luggage\", \"relation\": \"side_above\", \"distance_cm\": 18.4}, {\"object\": \"bag\", \"relation\": \"side_above\", \"distance_cm\": 36.8}]}", "target": "bottle", "split": "validation"}
{"id": "3799bd47b3_4", "scene_id": "3799bd47b3", "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": "validation"}
{"id": "3799bd47b3_5", "scene_id": "3799bd47b3", "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. guitar\nB. curtain\nC. heater\nD. windowframe", "choices": ["A. guitar", "B. curtain", "C. heater", "D. windowframe"], "answer": "A. guitar", "target": "cup", "split": "validation"}
{"id": "3799bd47b3_6", "scene_id": "3799bd47b3", "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 guitar, about 11 cm away. It is also to the side of and below the window, about 18 cm away. It is further to the side of and below the windowframe, about 19 cm away.", "target": "cup", "split": "validation"}
{"id": "3799bd47b3_7", "scene_id": "3799bd47b3", "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\": \"window\", \"references\": [{\"object\": \"guitar\", \"relation\": \"side_below\", \"distance_cm\": 10.7}, {\"object\": \"window\", \"relation\": \"side_below\", \"distance_cm\": 18.0}]}", "target": "cup", "split": "validation"}
{"id": "3799bd47b3_8", "scene_id": "3799bd47b3", "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": "validation"}
{"id": "3799bd47b3_9", "scene_id": "3799bd47b3", "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. lamp\nB. pedestal fan\nC. bookshelf\nD. power outlet", "choices": ["A. lamp", "B. pedestal fan", "C. bookshelf", "D. power outlet"], "answer": "A. lamp", "target": "power strip", "split": "validation"}
{"id": "3799bd47b3_10", "scene_id": "3799bd47b3", "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 lamp, about 7 cm away. It is also to the side of and below the pedestal fan, about 12 cm away. It is further to the side of and below the bookshelf, about 38 cm away.", "target": "power strip", "split": "validation"}
{"id": "3799bd47b3_11", "scene_id": "3799bd47b3", "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\": \"bookshelf\", \"references\": [{\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 7.0}, {\"object\": \"pedestal fan\", \"relation\": \"side_below\", \"distance_cm\": 12.0}]}", "target": "power strip", "split": "validation"}
{"id": "39705fbdf1_0", "scene_id": "39705fbdf1", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "39705fbdf1_1", "scene_id": "39705fbdf1", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. table\nC. projector\nD. book", "choices": ["A. chair", "B. table", "C. projector", "D. book"], "answer": "B. table", "target": "extension cord", "split": "validation"}
{"id": "39705fbdf1_2", "scene_id": "39705fbdf1", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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. It is also to the side of and above the chair, about 44 cm away. It is further near the projector, about 44 cm away.", "target": "extension cord", "split": "validation"}
{"id": "39705fbdf1_3", "scene_id": "39705fbdf1", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 43.6}, {\"object\": \"projector\", \"relation\": \"near\", \"distance_cm\": 43.9}]}", "target": "extension cord", "split": "validation"}
{"id": "39705fbdf1_4", "scene_id": "39705fbdf1", "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": "No", "target": "water bottle", "split": "validation"}
{"id": "39705fbdf1_5", "scene_id": "39705fbdf1", "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. telephone\nB. mouse\nC. trash can\nD. book", "choices": ["A. telephone", "B. mouse", "C. trash can", "D. book"], "answer": "D. book", "target": "keyboard", "split": "validation"}
{"id": "39705fbdf1_6", "scene_id": "39705fbdf1", "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 cabinet. It is also next to the book, about 3 cm away. It is further near the mouse, about 20 cm away.", "target": "keyboard", "split": "validation"}
{"id": "39705fbdf1_7", "scene_id": "39705fbdf1", "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\": \"cabinet\", \"references\": [{\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 3.3}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 19.8}]}", "target": "keyboard", "split": "validation"}
{"id": "39705fbdf1_8", "scene_id": "39705fbdf1", "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": "validation"}
{"id": "39705fbdf1_9", "scene_id": "39705fbdf1", "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. keyboard\nC. remote\nD. telephone", "choices": ["A. book", "B. keyboard", "C. remote", "D. telephone"], "answer": "A. book", "target": "mouse", "split": "validation"}
{"id": "39705fbdf1_10", "scene_id": "39705fbdf1", "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 cabinet. It is also next to the book, about 6 cm away. It is further next to the telephone, about 8 cm away.", "target": "mouse", "split": "validation"}
{"id": "39705fbdf1_11", "scene_id": "39705fbdf1", "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\": \"cabinet\", \"references\": [{\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 6.3}, {\"object\": \"telephone\", \"relation\": \"next_to\", \"distance_cm\": 7.6}]}", "target": "mouse", "split": "validation"}
{"id": "39f36da05b_0", "scene_id": "39f36da05b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "39f36da05b_1", "scene_id": "39f36da05b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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\nB. table\nC. cup\nD. monitor", "choices": ["A. mouse", "B. table", "C. cup", "D. monitor"], "answer": "C. cup", "target": "calculator", "split": "validation"}
{"id": "39f36da05b_2", "scene_id": "39f36da05b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 cup, about 3 cm away. It is further near the mouse, about 44 cm away.", "target": "calculator", "split": "validation"}
{"id": "39f36da05b_3", "scene_id": "39f36da05b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 3.3}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 44.1}]}", "target": "calculator", "split": "validation"}
{"id": "39f36da05b_4", "scene_id": "39f36da05b", "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": "validation"}
{"id": "39f36da05b_5", "scene_id": "39f36da05b", "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 cable duct, about 7 cm away. It is also next to the power socket, about 8 cm away. It is further to the side of and below the window sill, about 12 cm away.", "target": "headphones", "split": "validation"}
{"id": "39f36da05b_6", "scene_id": "39f36da05b", "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\": \"cable duct\", \"relation\": \"next_to\", \"distance_cm\": 6.5}, {\"object\": \"power socket\", \"relation\": \"next_to\", \"distance_cm\": 7.5}]}", "target": "headphones", "split": "validation"}
{"id": "39f36da05b_7", "scene_id": "39f36da05b", "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": "validation"}
{"id": "39f36da05b_8", "scene_id": "39f36da05b", "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. cup\nB. whiteboard marker\nC. monitor\nD. whiteboard eraser", "choices": ["A. cup", "B. whiteboard marker", "C. monitor", "D. whiteboard eraser"], "answer": "B. whiteboard marker", "target": "keyboard", "split": "validation"}
{"id": "39f36da05b_9", "scene_id": "39f36da05b", "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 whiteboard marker, about 4 cm away. It is also next to the cup, about 6 cm away. It is further to the side of and below the monitor, about 6 cm away.", "target": "keyboard", "split": "validation"}
{"id": "39f36da05b_10", "scene_id": "39f36da05b", "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\": \"whiteboard marker\", \"relation\": \"next_to\", \"distance_cm\": 4.4}, {\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 6.0}]}", "target": "keyboard", "split": "validation"}
{"id": "39f36da05b_11", "scene_id": "39f36da05b", "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 appear?", "choices": null, "answer": "Yes", "target": "pen", "split": "validation"}
{"id": "39f36da05b_12", "scene_id": "39f36da05b", "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. table\nB. headphones\nC. keyboard\nD. mouse", "choices": ["A. table", "B. headphones", "C. keyboard", "D. mouse"], "answer": "A. table", "target": "pen", "split": "validation"}
{"id": "39f36da05b_13", "scene_id": "39f36da05b", "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 near the headphones, about 24 cm away. It is also near the mouse, about 25 cm away. It is further near the keyboard, about 32 cm away.", "target": "pen", "split": "validation"}
{"id": "39f36da05b_14", "scene_id": "39f36da05b", "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\": \"headphones\", \"relation\": \"near\", \"distance_cm\": 24.0}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 24.6}]}", "target": "pen", "split": "validation"}
{"id": "3a3745a437_0", "scene_id": "3a3745a437", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "3a3745a437_1", "scene_id": "3a3745a437", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. wall clock\nB. audio interface\nC. bag\nD. toilet paper", "choices": ["A. wall clock", "B. audio interface", "C. bag", "D. toilet paper"], "answer": "A. wall clock", "target": "backpack", "split": "validation"}
{"id": "3a3745a437_2", "scene_id": "3a3745a437", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 below the wall clock, about 33 cm away. It is further to the side of and above the toilet paper, about 67 cm away.", "target": "backpack", "split": "validation"}
{"id": "3a3745a437_3", "scene_id": "3a3745a437", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"wall clock\", \"relation\": \"side_below\", \"distance_cm\": 33.0}, {\"object\": \"toilet paper\", \"relation\": \"side_above\", \"distance_cm\": 66.9}]}", "target": "backpack", "split": "validation"}
{"id": "3a3745a437_4", "scene_id": "3a3745a437", "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": "validation"}
{"id": "3a3745a437_5", "scene_id": "3a3745a437", "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 desk. 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 jacket, about 12 cm away.", "target": "headphones", "split": "validation"}
{"id": "3a3745a437_6", "scene_id": "3a3745a437", "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\": \"desk\", \"references\": [{\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 11.8}, {\"object\": \"jacket\", \"relation\": \"side_below\", \"distance_cm\": 11.9}]}", "target": "headphones", "split": "validation"}
{"id": "3a3745a437_7", "scene_id": "3a3745a437", "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": "No", "target": "notebook", "split": "validation"}
{"id": "3a3745a437_8", "scene_id": "3a3745a437", "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. bag\nB. mug\nC. shelf\nD. chair", "choices": ["A. bag", "B. mug", "C. shelf", "D. chair"], "answer": "B. mug", "target": "keyboard", "split": "validation"}
{"id": "3a3745a437_9", "scene_id": "3a3745a437", "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 next to the mug, about 7 cm away. It is further above the chair, about 9 cm away.", "target": "keyboard", "split": "validation"}
{"id": "3a3745a437_10", "scene_id": "3a3745a437", "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\": \"mug\", \"relation\": \"next_to\", \"distance_cm\": 6.9}, {\"object\": \"chair\", \"relation\": \"above\", \"distance_cm\": 9.0}]}", "target": "keyboard", "split": "validation"}
{"id": "3a3745a437_11", "scene_id": "3a3745a437", "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 pencil case appear?", "choices": null, "answer": "No", "target": "pencil case", "split": "validation"}
{"id": "3a3745a437_12", "scene_id": "3a3745a437", "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. bed\nB. pillow\nC. mattress\nD. blanket", "choices": ["A. bed", "B. pillow", "C. mattress", "D. blanket"], "answer": "C. mattress", "target": "phone charger", "split": "validation"}
{"id": "3a3745a437_13", "scene_id": "3a3745a437", "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 next to the mattress, about 3 cm away. It is also next to the blanket, about 10 cm away. It is further to the side of and below the pillow, about 35 cm away.", "target": "phone charger", "split": "validation"}
{"id": "3a3745a437_14", "scene_id": "3a3745a437", "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\": \"bed\", \"references\": [{\"object\": \"mattress\", \"relation\": \"next_to\", \"distance_cm\": 2.5}, {\"object\": \"blanket\", \"relation\": \"next_to\", \"distance_cm\": 9.7}]}", "target": "phone charger", "split": "validation"}
{"id": "3aa115e55e_0", "scene_id": "3aa115e55e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "3aa115e55e_1", "scene_id": "3aa115e55e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. cabinet\nB. power sockets\nC. speaker\nD. extension cord", "choices": ["A. cabinet", "B. power sockets", "C. speaker", "D. extension cord"], "answer": "A. cabinet", "target": "remote control", "split": "validation"}
{"id": "3aa115e55e_2", "scene_id": "3aa115e55e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 cabinet, about 20 cm away. It is further to the side of and below the speaker, about 39 cm away.", "target": "remote control", "split": "validation"}
{"id": "3aa115e55e_3", "scene_id": "3aa115e55e", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 19.8}, {\"object\": \"speaker\", \"relation\": \"side_below\", \"distance_cm\": 39.4}]}", "target": "remote control", "split": "validation"}
{"id": "3db0a1c8f3_0", "scene_id": "3db0a1c8f3", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "3db0a1c8f3_1", "scene_id": "3db0a1c8f3", "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 light switch, about 14 cm away. It is also near the beanbag, about 20 cm away. It is further to the side of and below the books, about 60 cm away.", "target": "backpack", "split": "validation"}
{"id": "3db0a1c8f3_2", "scene_id": "3db0a1c8f3", "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\": \"books\", \"references\": [{\"object\": \"light switch\", \"relation\": \"side_below\", \"distance_cm\": 13.9}, {\"object\": \"beanbag\", \"relation\": \"near\", \"distance_cm\": 20.5}]}", "target": "backpack", "split": "validation"}
{"id": "3db0a1c8f3_3", "scene_id": "3db0a1c8f3", "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": "validation"}
{"id": "3db0a1c8f3_4", "scene_id": "3db0a1c8f3", "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. pillar\nB. chair\nC. beanbag\nD. backpack", "choices": ["A. pillar", "B. chair", "C. beanbag", "D. backpack"], "answer": "A. pillar", "target": "bottle", "split": "validation"}
{"id": "3db0a1c8f3_5", "scene_id": "3db0a1c8f3", "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 on the table. It is also to the side of and below the pillar, about 42 cm away. It is further to the side of and above the chair, about 48 cm away.", "target": "bottle", "split": "validation"}
{"id": "3db0a1c8f3_6", "scene_id": "3db0a1c8f3", "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\": \"pillar\", \"relation\": \"side_below\", \"distance_cm\": 42.0}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 48.2}]}", "target": "bottle", "split": "validation"}
{"id": "3db0a1c8f3_7", "scene_id": "3db0a1c8f3", "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 lotion bottle appear?", "choices": null, "answer": "No", "target": "lotion bottle", "split": "validation"}
{"id": "3db0a1c8f3_8", "scene_id": "3db0a1c8f3", "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 headset in this scene?\nA. book\nB. piano chair\nC. backpack\nD. beanbag", "choices": ["A. book", "B. piano chair", "C. backpack", "D. beanbag"], "answer": "B. piano chair", "target": "headset", "split": "validation"}
{"id": "3db0a1c8f3_9", "scene_id": "3db0a1c8f3", "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 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 piano chair, about 36 cm away.", "target": "headset", "split": "validation"}
{"id": "3db0a1c8f3_10", "scene_id": "3db0a1c8f3", "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 headset in structured JSON format.", "choices": null, "answer": "{\"target\": \"headset\", \"support_surface\": \"piano chair\", \"references\": [{\"object\": \"piano chair\", \"relation\": \"side_above\", \"distance_cm\": 36.3}]}", "target": "headset", "split": "validation"}
{"id": "3db0a1c8f3_11", "scene_id": "3db0a1c8f3", "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 keyboard appear?", "choices": null, "answer": "Yes", "target": "keyboard", "split": "validation"}
{"id": "3db0a1c8f3_12", "scene_id": "3db0a1c8f3", "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. tv receiver\nB. light switch\nC. power socket\nD. shelf", "choices": ["A. tv receiver", "B. light switch", "C. power socket", "D. shelf"], "answer": "A. tv receiver", "target": "keyboard", "split": "validation"}
{"id": "3db0a1c8f3_13", "scene_id": "3db0a1c8f3", "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 near the tv receiver, about 18 cm away. It is also to the side of and below the light switch, about 33 cm away. It is further to the side of and below the power socket, about 39 cm away.", "target": "keyboard", "split": "validation"}
{"id": "3db0a1c8f3_14", "scene_id": "3db0a1c8f3", "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\": \"tv stand\", \"references\": [{\"object\": \"tv receiver\", \"relation\": \"near\", \"distance_cm\": 18.3}, {\"object\": \"light switch\", \"relation\": \"side_below\", \"distance_cm\": 33.4}]}", "target": "keyboard", "split": "validation"}
{"id": "3db0a1c8f3_15", "scene_id": "3db0a1c8f3", "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 laptop charger appear?", "choices": null, "answer": "No", "target": "laptop charger", "split": "validation"}
{"id": "3db0a1c8f3_16", "scene_id": "3db0a1c8f3", "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. power socket\nB. table\nC. carton\nD. book", "choices": ["A. power socket", "B. table", "C. carton", "D. book"], "answer": "D. book", "target": "mouse", "split": "validation"}
{"id": "3db0a1c8f3_17", "scene_id": "3db0a1c8f3", "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 book, about 7 cm away. It is also near the power socket, about 27 cm away. It is further to the side of and below the table, about 29 cm away.", "target": "mouse", "split": "validation"}
{"id": "3db0a1c8f3_18", "scene_id": "3db0a1c8f3", "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\": \"shelf\", \"references\": [{\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 6.6}, {\"object\": \"power socket\", \"relation\": \"near\", \"distance_cm\": 26.8}]}", "target": "mouse", "split": "validation"}
{"id": "3eba679830_0", "scene_id": "3eba679830", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "3eba679830_1", "scene_id": "3eba679830", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. file folder\nC. office chair\nD. tissue box", "choices": ["A. cabinet", "B. file folder", "C. office chair", "D. tissue box"], "answer": "A. cabinet", "target": "backpack", "split": "validation"}
{"id": "3eba679830_2", "scene_id": "3eba679830", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 tissue box, about 18 cm away. It is further to the side of and below the office chair, about 19 cm away.", "target": "backpack", "split": "validation"}
{"id": "3eba679830_3", "scene_id": "3eba679830", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 6.3}, {\"object\": \"tissue box\", \"relation\": \"side_below\", \"distance_cm\": 18.3}]}", "target": "backpack", "split": "validation"}
{"id": "3eba679830_4", "scene_id": "3eba679830", "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": "validation"}
{"id": "3eba679830_5", "scene_id": "3eba679830", "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. office chair\nD. cabinet", "choices": ["A. mouse", "B. monitor", "C. office chair", "D. cabinet"], "answer": "A. mouse", "target": "keyboard", "split": "validation"}
{"id": "3eba679830_6", "scene_id": "3eba679830", "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 near the mouse, about 16 cm away. It is further to the side of and below the monitor, about 29 cm away.", "target": "keyboard", "split": "validation"}
{"id": "3eba679830_7", "scene_id": "3eba679830", "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\", \"relation\": \"near\", \"distance_cm\": 15.6}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 29.1}]}", "target": "keyboard", "split": "validation"}
{"id": "3eba679830_8", "scene_id": "3eba679830", "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": "validation"}
{"id": "3eba679830_9", "scene_id": "3eba679830", "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. lamp\nB. curtain\nC. monitor\nD. keyboard", "choices": ["A. lamp", "B. curtain", "C. monitor", "D. keyboard"], "answer": "C. monitor", "target": "mouse", "split": "validation"}
{"id": "3eba679830_10", "scene_id": "3eba679830", "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 below the monitor, about 6 cm away. It is further near the keyboard, about 16 cm away.", "target": "mouse", "split": "validation"}
{"id": "3eba679830_11", "scene_id": "3eba679830", "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\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 6.5}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 15.6}]}", "target": "mouse", "split": "validation"}
{"id": "3eba679830_12", "scene_id": "3eba679830", "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": "validation"}
{"id": "3eba679830_13", "scene_id": "3eba679830", "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. backpack\nB. power outlet\nC. café curtains\nD. cabinet", "choices": ["A. backpack", "B. power outlet", "C. café curtains", "D. cabinet"], "answer": "B. power outlet", "target": "power strip", "split": "validation"}
{"id": "3eba679830_14", "scene_id": "3eba679830", "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 power outlet, about 48 cm away.", "target": "power strip", "split": "validation"}
{"id": "3eba679830_15", "scene_id": "3eba679830", "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\": \"vacuum cleaner\", \"references\": [{\"object\": \"power outlet\", \"relation\": \"side_below\", \"distance_cm\": 48.0}]}", "target": "power strip", "split": "validation"}
{"id": "3eba679830_16", "scene_id": "3eba679830", "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 comb appear?", "choices": null, "answer": "No", "target": "comb", "split": "validation"}
{"id": "3eba679830_17", "scene_id": "3eba679830", "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. office chair\nB. file folder\nC. backpack\nD. cabinet", "choices": ["A. office chair", "B. file folder", "C. backpack", "D. cabinet"], "answer": "B. file folder", "target": "tissue box", "split": "validation"}
{"id": "3eba679830_18", "scene_id": "3eba679830", "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 on the cabinet. It is also next to the file folder, about 6 cm away. It is further to the side of and above the backpack, about 18 cm away.", "target": "tissue box", "split": "validation"}
{"id": "3eba679830_19", "scene_id": "3eba679830", "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\": \"cabinet\", \"references\": [{\"object\": \"file folder\", \"relation\": \"next_to\", \"distance_cm\": 6.5}, {\"object\": \"backpack\", \"relation\": \"side_above\", \"distance_cm\": 18.3}]}", "target": "tissue box", "split": "validation"}
{"id": "3ee9d33aef_0", "scene_id": "3ee9d33aef", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "3ee9d33aef_1", "scene_id": "3ee9d33aef", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. nightstand\nB. clothes\nC. power strip\nD. power adapter", "choices": ["A. nightstand", "B. clothes", "C. power strip", "D. power adapter"], "answer": "C. power strip", "target": "cable", "split": "validation"}
{"id": "3ee9d33aef_2", "scene_id": "3ee9d33aef", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 next to the power strip, about 10 cm away. It is also to the side of and below the power adapter, about 14 cm away. It is further to the side of and below the nightstand, about 21 cm away.", "target": "cable", "split": "validation"}
{"id": "3ee9d33aef_3", "scene_id": "3ee9d33aef", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"nightstand\", \"references\": [{\"object\": \"power strip\", \"relation\": \"next_to\", \"distance_cm\": 10.0}, {\"object\": \"power adapter\", \"relation\": \"side_below\", \"distance_cm\": 13.7}]}", "target": "cable", "split": "validation"}
{"id": "3ee9d33aef_4", "scene_id": "3ee9d33aef", "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": "validation"}
{"id": "3ee9d33aef_5", "scene_id": "3ee9d33aef", "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. pen stand\nB. split\nC. bottle\nD. tissue box", "choices": ["A. pen stand", "B. split", "C. bottle", "D. tissue box"], "answer": "B. split", "target": "headphones", "split": "validation"}
{"id": "3ee9d33aef_6", "scene_id": "3ee9d33aef", "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 nightstand. It is also next to the split. It is further under the bottle, about 5 cm away.", "target": "headphones", "split": "validation"}
{"id": "3ee9d33aef_7", "scene_id": "3ee9d33aef", "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\": \"nightstand\", \"references\": [{\"object\": \"split\", \"relation\": \"next_to\", \"distance_cm\": 2.0}, {\"object\": \"bottle\", \"relation\": \"under\", \"distance_cm\": 5.0}]}", "target": "headphones", "split": "validation"}
{"id": "3ee9d33aef_8", "scene_id": "3ee9d33aef", "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": "validation"}
{"id": "3ee9d33aef_9", "scene_id": "3ee9d33aef", "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. cable\nB. office chair\nC. office table\nD. nightstand", "choices": ["A. cable", "B. office chair", "C. office table", "D. nightstand"], "answer": "A. cable", "target": "power strip", "split": "validation"}
{"id": "3ee9d33aef_10", "scene_id": "3ee9d33aef", "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 next to the cable, about 10 cm away. It is also to the side of and below the office table, about 12 cm away. It is further to the side of and below the office chair, about 26 cm away.", "target": "power strip", "split": "validation"}
{"id": "3ee9d33aef_11", "scene_id": "3ee9d33aef", "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\": \"office table\", \"references\": [{\"object\": \"cable\", \"relation\": \"next_to\", \"distance_cm\": 10.0}, {\"object\": \"office table\", \"relation\": \"side_below\", \"distance_cm\": 11.7}]}", "target": "power strip", "split": "validation"}
{"id": "3ee9d33aef_12", "scene_id": "3ee9d33aef", "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": "validation"}
{"id": "3ee9d33aef_13", "scene_id": "3ee9d33aef", "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. pen stand\nB. book\nC. split\nD. headphones", "choices": ["A. pen stand", "B. book", "C. split", "D. headphones"], "answer": "A. pen stand", "target": "tissue box", "split": "validation"}
{"id": "3ee9d33aef_14", "scene_id": "3ee9d33aef", "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 on the nightstand. It is also under the pen stand. It is further to the side of and above the book, about 4 cm away.", "target": "tissue box", "split": "validation"}
{"id": "3ee9d33aef_15", "scene_id": "3ee9d33aef", "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\": \"nightstand\", \"references\": [{\"object\": \"pen stand\", \"relation\": \"under\", \"distance_cm\": 2.1}, {\"object\": \"book\", \"relation\": \"side_above\", \"distance_cm\": 3.5}]}", "target": "tissue box", "split": "validation"}
{"id": "3f15a9266d_0", "scene_id": "3f15a9266d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "3f15a9266d_1", "scene_id": "3f15a9266d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. book\nC. folder\nD. chair", "choices": ["A. mouse", "B. book", "C. folder", "D. chair"], "answer": "A. mouse", "target": "keyboard", "split": "validation"}
{"id": "3f15a9266d_2", "scene_id": "3f15a9266d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 next to the book, about 10 cm away.", "target": "keyboard", "split": "validation"}
{"id": "3f15a9266d_3", "scene_id": "3f15a9266d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 8.3}, {\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 10.4}]}", "target": "keyboard", "split": "validation"}
{"id": "3f15a9266d_4", "scene_id": "3f15a9266d", "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": "validation"}
{"id": "3f15a9266d_5", "scene_id": "3f15a9266d", "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. monitor\nD. book", "choices": ["A. chair", "B. keyboard", "C. monitor", "D. book"], "answer": "B. keyboard", "target": "mouse", "split": "validation"}
{"id": "3f15a9266d_6", "scene_id": "3f15a9266d", "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 8 cm away. It is also to the side of and below the monitor, about 28 cm away. It is further to the side of and above the chair, about 30 cm away.", "target": "mouse", "split": "validation"}
{"id": "3f15a9266d_7", "scene_id": "3f15a9266d", "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\": \"chair\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 8.3}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 27.8}]}", "target": "mouse", "split": "validation"}
{"id": "3f15a9266d_8", "scene_id": "3f15a9266d", "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": "validation"}
{"id": "3f15a9266d_9", "scene_id": "3f15a9266d", "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. book\nB. folder\nC. chair\nD. monitor", "choices": ["A. book", "B. folder", "C. chair", "D. monitor"], "answer": "C. chair", "target": "mouse pad", "split": "validation"}
{"id": "3f15a9266d_10", "scene_id": "3f15a9266d", "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 chair, about 24 cm away. It is also to the side of and below the monitor, about 26 cm away. It is further near the book, about 50 cm away.", "target": "mouse pad", "split": "validation"}
{"id": "3f15a9266d_11", "scene_id": "3f15a9266d", "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\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 24.2}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 25.7}]}", "target": "mouse pad", "split": "validation"}
{"id": "3f15a9266d_12", "scene_id": "3f15a9266d", "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": "validation"}
{"id": "3f15a9266d_13", "scene_id": "3f15a9266d", "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. cup\nB. book\nC. telephone\nD. picture", "choices": ["A. cup", "B. book", "C. telephone", "D. picture"], "answer": "A. cup", "target": "pencil holder", "split": "validation"}
{"id": "3f15a9266d_14", "scene_id": "3f15a9266d", "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 next to the cup, about 3 cm away. It is also to the side of and below the window, about 7 cm away. It is further to the side of and below the picture, about 10 cm away.", "target": "pencil holder", "split": "validation"}
{"id": "3f15a9266d_15", "scene_id": "3f15a9266d", "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\": \"window\", \"references\": [{\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 3.4}, {\"object\": \"window\", \"relation\": \"side_below\", \"distance_cm\": 7.4}]}", "target": "pencil holder", "split": "validation"}
{"id": "3f1e1610de_0", "scene_id": "3f1e1610de", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "3f1e1610de_1", "scene_id": "3f1e1610de", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. exercise ball\nB. speaker\nC. monitor\nD. mouse", "choices": ["A. exercise ball", "B. speaker", "C. monitor", "D. mouse"], "answer": "A. exercise ball", "target": "keyboard", "split": "validation"}
{"id": "3f1e1610de_2", "scene_id": "3f1e1610de", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 exercise ball, about 7 cm away. It is further next to the mouse, about 10 cm away.", "target": "keyboard", "split": "validation"}
{"id": "3f1e1610de_3", "scene_id": "3f1e1610de", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"exercise ball\", \"relation\": \"side_above\", \"distance_cm\": 6.6}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 9.6}]}", "target": "keyboard", "split": "validation"}
{"id": "3f1e1610de_4", "scene_id": "3f1e1610de", "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": "validation"}
{"id": "3f1e1610de_5", "scene_id": "3f1e1610de", "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. cabinet\nB. exercise ball\nC. desk\nD. keyboard", "choices": ["A. cabinet", "B. exercise ball", "C. desk", "D. keyboard"], "answer": "C. desk", "target": "mouse", "split": "validation"}
{"id": "3f1e1610de_6", "scene_id": "3f1e1610de", "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 8 cm away. It is also next to the keyboard, about 10 cm away. It is further to the side of and above the cabinet, about 19 cm away.", "target": "mouse", "split": "validation"}
{"id": "3f1e1610de_7", "scene_id": "3f1e1610de", "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\": 7.9}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 9.6}]}", "target": "mouse", "split": "validation"}
{"id": "3f1e1610de_8", "scene_id": "3f1e1610de", "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": "validation"}
{"id": "3f1e1610de_9", "scene_id": "3f1e1610de", "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. exercise ball\nB. speaker\nC. monitor\nD. cabinet", "choices": ["A. exercise ball", "B. speaker", "C. monitor", "D. cabinet"], "answer": "D. cabinet", "target": "mouse pad", "split": "validation"}
{"id": "3f1e1610de_10", "scene_id": "3f1e1610de", "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 to the side of and above the cabinet, about 14 cm away. It is further to the side of and above the exercise ball, about 18 cm away.", "target": "mouse pad", "split": "validation"}
{"id": "3f1e1610de_11", "scene_id": "3f1e1610de", "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\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 14.0}, {\"object\": \"exercise ball\", \"relation\": \"side_above\", \"distance_cm\": 18.1}]}", "target": "mouse pad", "split": "validation"}
{"id": "40b56bf310_0", "scene_id": "40b56bf310", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "40b56bf310_1", "scene_id": "40b56bf310", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. socket\nC. column\nD. monitor", "choices": ["A. cabinet", "B. socket", "C. column", "D. monitor"], "answer": "D. monitor", "target": "mouse", "split": "validation"}
{"id": "40b56bf310_2", "scene_id": "40b56bf310", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 18 cm away. It is further near the socket, about 19 cm away.", "target": "mouse", "split": "validation"}
{"id": "40b56bf310_3", "scene_id": "40b56bf310", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 17.8}, {\"object\": \"socket\", \"relation\": \"near\", \"distance_cm\": 19.2}]}", "target": "mouse", "split": "validation"}
{"id": "413085a827_0", "scene_id": "413085a827", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "413085a827_1", "scene_id": "413085a827", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. jar\nB. office chair\nC. plate\nD. mug", "choices": ["A. jar", "B. office chair", "C. plate", "D. mug"], "answer": "D. mug", "target": "bottle", "split": "validation"}
{"id": "413085a827_2", "scene_id": "413085a827", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 office chair, about 13 cm away. It is further to the side of and above the plate, about 27 cm away.", "target": "bottle", "split": "validation"}
{"id": "413085a827_3", "scene_id": "413085a827", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 13.4}, {\"object\": \"plate\", \"relation\": \"side_above\", \"distance_cm\": 26.5}]}", "target": "bottle", "split": "validation"}
{"id": "413085a827_4", "scene_id": "413085a827", "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": "validation"}
{"id": "413085a827_5", "scene_id": "413085a827", "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. jackets\nB. bottle\nC. puch\nD. paper bag", "choices": ["A. jackets", "B. bottle", "C. puch", "D. paper bag"], "answer": "B. bottle", "target": "mug", "split": "validation"}
{"id": "413085a827_6", "scene_id": "413085a827", "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 under the bottle, about 3 cm away. It is also near the puch, about 24 cm away. It is further to the side of and above the jackets, about 31 cm away.", "target": "mug", "split": "validation"}
{"id": "413085a827_7", "scene_id": "413085a827", "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\": \"bag\", \"references\": [{\"object\": \"bottle\", \"relation\": \"under\", \"distance_cm\": 3.0}, {\"object\": \"puch\", \"relation\": \"near\", \"distance_cm\": 24.4}]}", "target": "mug", "split": "validation"}
{"id": "413085a827_8", "scene_id": "413085a827", "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": "validation"}
{"id": "413085a827_9", "scene_id": "413085a827", "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 extension in this scene?\nA. heater\nB. table\nC. power socket\nD. electric socket", "choices": ["A. heater", "B. table", "C. power socket", "D. electric socket"], "answer": "B. table", "target": "power extension", "split": "validation"}
{"id": "413085a827_10", "scene_id": "413085a827", "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 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 21 cm away. It is also to the side of and below the power socket, about 32 cm away. It is further to the side of and below the heater, about 52 cm away.", "target": "power extension", "split": "validation"}
{"id": "413085a827_11", "scene_id": "413085a827", "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 extension in structured JSON format.", "choices": null, "answer": "{\"target\": \"power extension\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 20.8}, {\"object\": \"power socket\", \"relation\": \"side_below\", \"distance_cm\": 31.7}]}", "target": "power extension", "split": "validation"}
{"id": "43cd995c51_0", "scene_id": "43cd995c51", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "43cd995c51_1", "scene_id": "43cd995c51", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. tray\nB. books\nC. table\nD. desk lamp", "choices": ["A. tray", "B. books", "C. table", "D. desk lamp"], "answer": "C. table", "target": "headphones", "split": "validation"}
{"id": "43cd995c51_2", "scene_id": "43cd995c51", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 22 cm away. It is also to the side of and above the tray, about 44 cm away. It is further to the side of and above the books, about 53 cm away.", "target": "headphones", "split": "validation"}
{"id": "43cd995c51_3", "scene_id": "43cd995c51", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 22.4}, {\"object\": \"tray\", \"relation\": \"side_above\", \"distance_cm\": 43.8}]}", "target": "headphones", "split": "validation"}
{"id": "4422722c49_0", "scene_id": "4422722c49", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "4422722c49_1", "scene_id": "4422722c49", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. power strip\nB. box\nC. wifi router\nD. power socket", "choices": ["A. power strip", "B. box", "C. wifi router", "D. power socket"], "answer": "B. box", "target": "charger", "split": "validation"}
{"id": "4422722c49_2", "scene_id": "4422722c49", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 box, about 7 cm away. It is further next to the power strip, about 8 cm away.", "target": "charger", "split": "validation"}
{"id": "4422722c49_3", "scene_id": "4422722c49", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 7.1}, {\"object\": \"power strip\", \"relation\": \"next_to\", \"distance_cm\": 8.2}]}", "target": "charger", "split": "validation"}
{"id": "4422722c49_4", "scene_id": "4422722c49", "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": "validation"}
{"id": "4422722c49_5", "scene_id": "4422722c49", "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. shelf\nC. book\nD. file binder", "choices": ["A. box", "B. shelf", "C. book", "D. file binder"], "answer": "C. book", "target": "cup", "split": "validation"}
{"id": "4422722c49_6", "scene_id": "4422722c49", "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 file binder, about 10 cm away. It is also to the side of and below the box, about 40 cm away. It is further near the paper towel dispenser, about 50 cm away.", "target": "cup", "split": "validation"}
{"id": "4422722c49_7", "scene_id": "4422722c49", "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\": \"file binder\", \"relation\": \"side_below\", \"distance_cm\": 9.8}, {\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 39.7}]}", "target": "cup", "split": "validation"}
{"id": "4422722c49_8", "scene_id": "4422722c49", "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": "validation"}
{"id": "4422722c49_9", "scene_id": "4422722c49", "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 keyboard, about 11 cm away. It is further next to the box, about 11 cm away.", "target": "mouse", "split": "validation"}
{"id": "4422722c49_10", "scene_id": "4422722c49", "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\": 10.5}, {\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 10.8}]}", "target": "mouse", "split": "validation"}
{"id": "4422722c49_11", "scene_id": "4422722c49", "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 mug appear?", "choices": null, "answer": "No", "target": "mug", "split": "validation"}
{"id": "4422722c49_12", "scene_id": "4422722c49", "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. computer tower\nB. folder\nC. keyboard\nD. monitor", "choices": ["A. computer tower", "B. folder", "C. keyboard", "D. monitor"], "answer": "A. computer tower", "target": "power strip", "split": "validation"}
{"id": "4422722c49_13", "scene_id": "4422722c49", "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 table. It is also to the side of and below the computer tower, about 9 cm away. It is further to the side of and below the monitor, about 25 cm away.", "target": "power strip", "split": "validation"}
{"id": "4422722c49_14", "scene_id": "4422722c49", "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\": \"table\", \"references\": [{\"object\": \"computer tower\", \"relation\": \"side_below\", \"distance_cm\": 9.3}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 25.1}]}", "target": "power strip", "split": "validation"}
{"id": "4422722c49_15", "scene_id": "4422722c49", "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 box appear?", "choices": null, "answer": "Yes", "target": "tissue box", "split": "validation"}
{"id": "4422722c49_16", "scene_id": "4422722c49", "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. laboratory pellet press\nB. office chair\nC. step stool\nD. table", "choices": ["A. laboratory pellet press", "B. office chair", "C. step stool", "D. table"], "answer": "D. table", "target": "tissue box", "split": "validation"}
{"id": "4422722c49_17", "scene_id": "4422722c49", "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 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 below the laboratory pellet press, about 28 cm away.", "target": "tissue box", "split": "validation"}
{"id": "4422722c49_18", "scene_id": "4422722c49", "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\": \"table\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 17.6}, {\"object\": \"laboratory pellet press\", \"relation\": \"side_below\", \"distance_cm\": 27.9}]}", "target": "tissue box", "split": "validation"}
{"id": "4423a61d09_0", "scene_id": "4423a61d09", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "4423a61d09_1", "scene_id": "4423a61d09", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. telephone\nD. mouse pad", "choices": ["A. table", "B. monitor", "C. telephone", "D. mouse pad"], "answer": "A. table", "target": "headphones", "split": "validation"}
{"id": "4423a61d09_2", "scene_id": "4423a61d09", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 telephone, about 3 cm away. It is further next to the mouse pad, about 13 cm away.", "target": "headphones", "split": "validation"}
{"id": "4423a61d09_3", "scene_id": "4423a61d09", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 2.4}, {\"object\": \"telephone\", \"relation\": \"next_to\", \"distance_cm\": 3.5}]}", "target": "headphones", "split": "validation"}
{"id": "4423a61d09_4", "scene_id": "4423a61d09", "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": "validation"}
{"id": "4423a61d09_5", "scene_id": "4423a61d09", "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 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 file binder, about 6 cm away. It is further to the side of and below the monitor, about 6 cm away.", "target": "mouse pad", "split": "validation"}
{"id": "4423a61d09_6", "scene_id": "4423a61d09", "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 pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"file binder\", \"relation\": \"next_to\", \"distance_cm\": 6.0}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 6.4}]}", "target": "mouse pad", "split": "validation"}
{"id": "4423a61d09_7", "scene_id": "4423a61d09", "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": "validation"}
{"id": "4423a61d09_8", "scene_id": "4423a61d09", "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 power strip in this scene?\nA. monitor\nB. laptop\nC. cardboard box\nD. table fan", "choices": ["A. monitor", "B. laptop", "C. cardboard box", "D. table fan"], "answer": "A. monitor", "target": "power strip", "split": "validation"}
{"id": "4423a61d09_9", "scene_id": "4423a61d09", "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 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 monitor, about 16 cm away. It is also to the side of and below the table fan, about 22 cm away. It is further to the side of and above the cardboard box, about 24 cm away.", "target": "power strip", "split": "validation"}
{"id": "4423a61d09_10", "scene_id": "4423a61d09", "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 power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 15.6}, {\"object\": \"table fan\", \"relation\": \"side_below\", \"distance_cm\": 21.6}]}", "target": "power strip", "split": "validation"}
{"id": "442b144761_0", "scene_id": "442b144761", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "442b144761_1", "scene_id": "442b144761", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. markers\nB. chair\nC. power sockets\nD. whiteboard eraser", "choices": ["A. markers", "B. chair", "C. power sockets", "D. whiteboard eraser"], "answer": "C. power sockets", "target": "bottle", "split": "validation"}
{"id": "442b144761_2", "scene_id": "442b144761", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 power sockets, about 5 cm away. It is also to the side of and above the markers, about 33 cm away. It is further to the side of and above the whiteboard eraser, about 50 cm away.", "target": "bottle", "split": "validation"}
{"id": "442b144761_3", "scene_id": "442b144761", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"conference table\", \"references\": [{\"object\": \"power sockets\", \"relation\": \"next_to\", \"distance_cm\": 5.2}, {\"object\": \"markers\", \"relation\": \"side_above\", \"distance_cm\": 33.2}]}", "target": "bottle", "split": "validation"}
{"id": "469f112e38_0", "scene_id": "469f112e38", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "469f112e38_1", "scene_id": "469f112e38", "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 next to the frying pan, about 13 cm away. It is also next to the blender, about 13 cm away. It is further near the bowl, about 15 cm away.", "target": "spray bottle", "split": "validation"}
{"id": "469f112e38_2", "scene_id": "469f112e38", "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\": \"window\", \"references\": [{\"object\": \"frying pan\", \"relation\": \"next_to\", \"distance_cm\": 12.7}, {\"object\": \"blender\", \"relation\": \"next_to\", \"distance_cm\": 12.8}]}", "target": "spray bottle", "split": "validation"}
{"id": "469f112e38_3", "scene_id": "469f112e38", "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 pencil holder appear?", "choices": null, "answer": "No", "target": "pencil holder", "split": "validation"}
{"id": "469f112e38_4", "scene_id": "469f112e38", "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 tote bag in this scene?\nA. blender\nB. chair\nC. bottle\nD. bowl", "choices": ["A. blender", "B. chair", "C. bottle", "D. bowl"], "answer": "B. chair", "target": "tote bag", "split": "validation"}
{"id": "469f112e38_5", "scene_id": "469f112e38", "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 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 chair, about 65 cm away.", "target": "tote bag", "split": "validation"}
{"id": "469f112e38_6", "scene_id": "469f112e38", "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 tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"office chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 65.3}]}", "target": "tote bag", "split": "validation"}
{"id": "47c1209b9e_0", "scene_id": "47c1209b9e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "47c1209b9e_1", "scene_id": "47c1209b9e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. books\nC. router\nD. shelf", "choices": ["A. bowl", "B. books", "C. router", "D. shelf"], "answer": "D. shelf", "target": "backpack", "split": "validation"}
{"id": "47c1209b9e_2", "scene_id": "47c1209b9e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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, about 15 cm away. It is also to the side of and below the books, about 25 cm away. It is further to the side of and below the bowl, about 46 cm away.", "target": "backpack", "split": "validation"}
{"id": "47c1209b9e_3", "scene_id": "47c1209b9e", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 15.2}, {\"object\": \"books\", \"relation\": \"side_below\", \"distance_cm\": 24.7}]}", "target": "backpack", "split": "validation"}
{"id": "47c1209b9e_4", "scene_id": "47c1209b9e", "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": "validation"}
{"id": "47c1209b9e_5", "scene_id": "47c1209b9e", "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. mouse\nB. book\nC. mouse pad\nD. table lamp", "choices": ["A. mouse", "B. book", "C. mouse pad", "D. table lamp"], "answer": "C. mouse pad", "target": "laptop charger", "split": "validation"}
{"id": "47c1209b9e_6", "scene_id": "47c1209b9e", "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 on the desk. It is also next to the mouse pad, about 4 cm away. It is further next to the book, about 9 cm away.", "target": "laptop charger", "split": "validation"}
{"id": "47c1209b9e_7", "scene_id": "47c1209b9e", "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\": \"desk\", \"references\": [{\"object\": \"mouse pad\", \"relation\": \"next_to\", \"distance_cm\": 4.1}, {\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 8.8}]}", "target": "laptop charger", "split": "validation"}
{"id": "47c1209b9e_8", "scene_id": "47c1209b9e", "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": "validation"}
{"id": "47c1209b9e_9", "scene_id": "47c1209b9e", "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 desk. It is also next to the laptop charger, about 9 cm away. It is further to the side of and below the monitor, about 10 cm away.", "target": "mouse", "split": "validation"}
{"id": "47c1209b9e_10", "scene_id": "47c1209b9e", "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 charger\", \"relation\": \"next_to\", \"distance_cm\": 9.3}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 10.0}]}", "target": "mouse", "split": "validation"}
{"id": "47c1209b9e_11", "scene_id": "47c1209b9e", "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 pad appear?", "choices": null, "answer": "Yes", "target": "mouse pad", "split": "validation"}
{"id": "47c1209b9e_12", "scene_id": "47c1209b9e", "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 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 laptop charger, about 4 cm away. It is further under the monitor, about 5 cm away.", "target": "mouse pad", "split": "validation"}
{"id": "47c1209b9e_13", "scene_id": "47c1209b9e", "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 pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"laptop charger\", \"relation\": \"next_to\", \"distance_cm\": 4.1}, {\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 5.0}]}", "target": "mouse pad", "split": "validation"}
{"id": "47c1209b9e_14", "scene_id": "47c1209b9e", "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 bottle appear?", "choices": null, "answer": "No", "target": "bottle", "split": "validation"}
{"id": "47c1209b9e_15", "scene_id": "47c1209b9e", "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 tissue box in this scene?\nA. power socket\nB. shelf rail\nC. jar\nD. mayonnaise bottle", "choices": ["A. power socket", "B. shelf rail", "C. jar", "D. mayonnaise bottle"], "answer": "B. shelf rail", "target": "tissue box", "split": "validation"}
{"id": "47c1209b9e_16", "scene_id": "47c1209b9e", "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 tissue box in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue box is under the shelf rail. It is also to the side of and above the power socket, about 9 cm away. It is further to the side of and below the mayonnaise bottle, about 13 cm away.", "target": "tissue box", "split": "validation"}
{"id": "47c1209b9e_17", "scene_id": "47c1209b9e", "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 tissue box in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue box\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"shelf rail\", \"relation\": \"under\", \"distance_cm\": 2.4}, {\"object\": \"power socket\", \"relation\": \"side_above\", \"distance_cm\": 8.9}]}", "target": "tissue box", "split": "validation"}
{"id": "47c1209b9e_18", "scene_id": "47c1209b9e", "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 water bottle appear?", "choices": null, "answer": "Yes", "target": "water bottle", "split": "validation"}
{"id": "47c1209b9e_19", "scene_id": "47c1209b9e", "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 water bottle in this scene?\nA. bedside table\nB. mayonnaise bottle\nC. microwave\nD. jar", "choices": ["A. bedside table", "B. mayonnaise bottle", "C. microwave", "D. jar"], "answer": "C. microwave", "target": "water bottle", "split": "validation"}
{"id": "47c1209b9e_20", "scene_id": "47c1209b9e", "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 water bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The water bottle is next to the microwave, about 5 cm away. It is also to the side of and above the jar, about 35 cm away. It is further to the side of and above the mayonnaise bottle, about 36 cm away.", "target": "water bottle", "split": "validation"}
{"id": "47c1209b9e_21", "scene_id": "47c1209b9e", "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 water bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"water bottle\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"microwave\", \"relation\": \"next_to\", \"distance_cm\": 4.5}, {\"object\": \"jar\", \"relation\": \"side_above\", \"distance_cm\": 35.3}]}", "target": "water bottle", "split": "validation"}
{"id": "48573f4c95_0", "scene_id": "48573f4c95", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "48573f4c95_1", "scene_id": "48573f4c95", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. stapler\nC. pens\nD. paper", "choices": ["A. office table", "B. stapler", "C. pens", "D. paper"], "answer": "C. pens", "target": "cup", "split": "validation"}
{"id": "48573f4c95_2", "scene_id": "48573f4c95", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 office table, about 6 cm away. It is also next to the stapler, about 8 cm away. It is further to the side of and above the paper, about 14 cm away.", "target": "cup", "split": "validation"}
{"id": "48573f4c95_3", "scene_id": "48573f4c95", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"pen stand\", \"references\": [{\"object\": \"office table\", \"relation\": \"side_above\", \"distance_cm\": 5.6}, {\"object\": \"stapler\", \"relation\": \"next_to\", \"distance_cm\": 8.4}]}", "target": "cup", "split": "validation"}
{"id": "48701abb21_0", "scene_id": "48701abb21", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "48701abb21_1", "scene_id": "48701abb21", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. cable\nB. cooking pan\nC. cooking pot\nD. cabinet", "choices": ["A. cable", "B. cooking pan", "C. cooking pot", "D. cabinet"], "answer": "D. cabinet", "target": "bottle", "split": "validation"}
{"id": "48701abb21_2", "scene_id": "48701abb21", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 cabinet, about 16 cm away. It is also near the cooking pot, about 18 cm away. It is further to the side of and above the cooking pan, about 30 cm away.", "target": "bottle", "split": "validation"}
{"id": "48701abb21_3", "scene_id": "48701abb21", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"near\", \"distance_cm\": 15.8}, {\"object\": \"cooking pot\", \"relation\": \"near\", \"distance_cm\": 17.9}]}", "target": "bottle", "split": "validation"}
{"id": "48701abb21_4", "scene_id": "48701abb21", "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": "validation"}
{"id": "48701abb21_5", "scene_id": "48701abb21", "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. cooking pan\nB. bottle\nC. cabinet\nD. ceiling lamp", "choices": ["A. cooking pan", "B. bottle", "C. cabinet", "D. ceiling lamp"], "answer": "D. ceiling lamp", "target": "cable", "split": "validation"}
{"id": "48701abb21_6", "scene_id": "48701abb21", "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 ceiling lamp.", "target": "cable", "split": "validation"}
{"id": "48701abb21_7", "scene_id": "48701abb21", "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\": \"ceiling lamp\", \"references\": [{\"object\": \"ceiling lamp\", \"relation\": \"above\", \"distance_cm\": 2.4}]}", "target": "cable", "split": "validation"}
{"id": "48701abb21_8", "scene_id": "48701abb21", "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": "validation"}
{"id": "48701abb21_9", "scene_id": "48701abb21", "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. kettle\nB. knife\nC. kitchen cabinet\nD. water jug", "choices": ["A. kettle", "B. knife", "C. kitchen cabinet", "D. water jug"], "answer": "B. knife", "target": "spray bottle", "split": "validation"}
{"id": "48701abb21_10", "scene_id": "48701abb21", "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 to the side of and below the knife, about 5 cm away. It is also near the water jug, about 28 cm away. It is further to the side of and below the kitchen cabinet, about 39 cm away.", "target": "spray bottle", "split": "validation"}
{"id": "48701abb21_11", "scene_id": "48701abb21", "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\": \"knife\", \"relation\": \"side_below\", \"distance_cm\": 5.5}, {\"object\": \"water jug\", \"relation\": \"near\", \"distance_cm\": 28.5}]}", "target": "spray bottle", "split": "validation"}
{"id": "48da625e8c_0", "scene_id": "48da625e8c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "48da625e8c_1", "scene_id": "48da625e8c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. mouse\nC. tablet stand\nD. table", "choices": ["A. book", "B. mouse", "C. tablet stand", "D. table"], "answer": "D. table", "target": "bottle", "split": "validation"}
{"id": "48da625e8c_2", "scene_id": "48da625e8c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 below the tablet stand, about 12 cm away. It is further near the book, about 16 cm away.", "target": "bottle", "split": "validation"}
{"id": "48da625e8c_3", "scene_id": "48da625e8c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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 lamp\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 7.6}, {\"object\": \"tablet stand\", \"relation\": \"side_below\", \"distance_cm\": 11.8}]}", "target": "bottle", "split": "validation"}
{"id": "48da625e8c_4", "scene_id": "48da625e8c", "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": "validation"}
{"id": "48da625e8c_5", "scene_id": "48da625e8c", "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. plant\nC. box\nD. tablet stand", "choices": ["A. book", "B. plant", "C. box", "D. tablet stand"], "answer": "A. book", "target": "keyboard", "split": "validation"}
{"id": "48da625e8c_6", "scene_id": "48da625e8c", "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 box, about 7 cm away. It is further to the side of and below the plant, about 18 cm away.", "target": "keyboard", "split": "validation"}
{"id": "48da625e8c_7", "scene_id": "48da625e8c", "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\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 6.9}, {\"object\": \"plant\", \"relation\": \"side_below\", \"distance_cm\": 18.1}]}", "target": "keyboard", "split": "validation"}
{"id": "48da625e8c_8", "scene_id": "48da625e8c", "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": "validation"}
{"id": "48da625e8c_9", "scene_id": "48da625e8c", "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. tablet stand\nB. desk lamp\nC. box\nD. book", "choices": ["A. tablet stand", "B. desk lamp", "C. box", "D. book"], "answer": "C. box", "target": "mouse", "split": "validation"}
{"id": "48da625e8c_10", "scene_id": "48da625e8c", "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 box, about 5 cm away. It is further to the side of and below the desk lamp, about 7 cm away.", "target": "mouse", "split": "validation"}
{"id": "48da625e8c_11", "scene_id": "48da625e8c", "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\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 4.5}, {\"object\": \"desk lamp\", \"relation\": \"side_below\", \"distance_cm\": 6.8}]}", "target": "mouse", "split": "validation"}
{"id": "49a82360aa_0", "scene_id": "49a82360aa", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "49a82360aa_1", "scene_id": "49a82360aa", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. phone stand\nC. windowsill\nD. heater", "choices": ["A. bottle", "B. phone stand", "C. windowsill", "D. heater"], "answer": "B. phone stand", "target": "cup", "split": "validation"}
{"id": "49a82360aa_2", "scene_id": "49a82360aa", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 phone stand, about 7 cm away. It is further near the windowsill, about 21 cm away.", "target": "cup", "split": "validation"}
{"id": "49a82360aa_3", "scene_id": "49a82360aa", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"phone stand\", \"relation\": \"next_to\", \"distance_cm\": 6.8}, {\"object\": \"windowsill\", \"relation\": \"near\", \"distance_cm\": 21.0}]}", "target": "cup", "split": "validation"}
{"id": "49a82360aa_4", "scene_id": "49a82360aa", "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": "validation"}
{"id": "49a82360aa_5", "scene_id": "49a82360aa", "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 5 cm away. It is further next to the laptop, about 6 cm away.", "target": "mouse", "split": "validation"}
{"id": "49a82360aa_6", "scene_id": "49a82360aa", "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\": 5.5}, {\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 5.7}]}", "target": "mouse", "split": "validation"}
{"id": "49a82360aa_7", "scene_id": "49a82360aa", "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 controller appear?", "choices": null, "answer": "No", "target": "remote controller", "split": "validation"}
{"id": "49a82360aa_8", "scene_id": "49a82360aa", "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 stand in this scene?\nA. windowsill\nB. paper towel\nC. cup\nD. heater", "choices": ["A. windowsill", "B. paper towel", "C. cup", "D. heater"], "answer": "B. paper towel", "target": "phone stand", "split": "validation"}
{"id": "49a82360aa_9", "scene_id": "49a82360aa", "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 stand in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone stand is on the table. It is also next to the paper towel, about 3 cm away. It is further next to the cup, about 7 cm away.", "target": "phone stand", "split": "validation"}
{"id": "49a82360aa_10", "scene_id": "49a82360aa", "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 stand in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone stand\", \"support_surface\": \"table\", \"references\": [{\"object\": \"paper towel\", \"relation\": \"next_to\", \"distance_cm\": 3.0}, {\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 6.8}]}", "target": "phone stand", "split": "validation"}
{"id": "49a82360aa_11", "scene_id": "49a82360aa", "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 strip appear?", "choices": null, "answer": "Yes", "target": "power strip", "split": "validation"}
{"id": "49a82360aa_12", "scene_id": "49a82360aa", "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. tea pot\nB. tablet\nC. basket\nD. tv stand", "choices": ["A. tea pot", "B. tablet", "C. basket", "D. tv stand"], "answer": "C. basket", "target": "power strip", "split": "validation"}
{"id": "49a82360aa_13", "scene_id": "49a82360aa", "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 to the side of and above the basket, about 15 cm away. It is also near the tablet, about 19 cm away. It is further to the side of and below the tea pot, about 26 cm away.", "target": "power strip", "split": "validation"}
{"id": "49a82360aa_14", "scene_id": "49a82360aa", "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\": \"bookshelf\", \"references\": [{\"object\": \"basket\", \"relation\": \"side_above\", \"distance_cm\": 15.1}, {\"object\": \"tablet\", \"relation\": \"near\", \"distance_cm\": 18.9}]}", "target": "power strip", "split": "validation"}
{"id": "54b6127146_0", "scene_id": "54b6127146", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "54b6127146_1", "scene_id": "54b6127146", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. telephone\nB. case\nC. nightstand\nD. suitcase", "choices": ["A. telephone", "B. case", "C. nightstand", "D. suitcase"], "answer": "D. suitcase", "target": "power strip", "split": "validation"}
{"id": "54b6127146_2", "scene_id": "54b6127146", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 suitcase, about 43 cm away. It is also to the side of and below the nightstand, about 54 cm away. It is further to the side of and below the telephone, about 63 cm away.", "target": "power strip", "split": "validation"}
{"id": "54b6127146_3", "scene_id": "54b6127146", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"suitcase\", \"relation\": \"side_below\", \"distance_cm\": 43.1}, {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 54.2}]}", "target": "power strip", "split": "validation"}
{"id": "54b6127146_4", "scene_id": "54b6127146", "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": "validation"}
{"id": "54b6127146_5", "scene_id": "54b6127146", "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. painting\nB. towel\nC. sink\nD. toilet", "choices": ["A. painting", "B. towel", "C. sink", "D. toilet"], "answer": "B. towel", "target": "tissue box", "split": "validation"}
{"id": "54b6127146_6", "scene_id": "54b6127146", "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 counter. It is also to the side of and above the towel, about 15 cm away. It is further to the side of and above the toilet, about 29 cm away.", "target": "tissue box", "split": "validation"}
{"id": "54b6127146_7", "scene_id": "54b6127146", "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\": \"counter\", \"references\": [{\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 15.3}, {\"object\": \"toilet\", \"relation\": \"side_above\", \"distance_cm\": 29.1}]}", "target": "tissue box", "split": "validation"}
{"id": "5748ce6f01_0", "scene_id": "5748ce6f01", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "5748ce6f01_1", "scene_id": "5748ce6f01", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. table\nD. keyboard", "choices": ["A. office chair", "B. monitor", "C. table", "D. keyboard"], "answer": "A. office chair", "target": "cup", "split": "validation"}
{"id": "5748ce6f01_2", "scene_id": "5748ce6f01", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 office chair, about 50 cm away.", "target": "cup", "split": "validation"}
{"id": "5748ce6f01_3", "scene_id": "5748ce6f01", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 50.0}]}", "target": "cup", "split": "validation"}
{"id": "57d9fecae9_0", "scene_id": "57d9fecae9", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "57d9fecae9_1", "scene_id": "57d9fecae9", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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 strip\nB. hanging storage\nC. basket\nD. coax outlet", "choices": ["A. power strip", "B. hanging storage", "C. basket", "D. coax outlet"], "answer": "A. power strip", "target": "cable", "split": "validation"}
{"id": "57d9fecae9_2", "scene_id": "57d9fecae9", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 next to the power strip, about 3 cm away. It is also to the side of and below the hanging storage, about 9 cm away. It is further to the side of and below the coax outlet, about 10 cm away.", "target": "cable", "split": "validation"}
{"id": "57d9fecae9_3", "scene_id": "57d9fecae9", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"power strip\", \"relation\": \"next_to\", \"distance_cm\": 3.0}, {\"object\": \"hanging storage\", \"relation\": \"side_below\", \"distance_cm\": 9.4}]}", "target": "cable", "split": "validation"}
{"id": "57d9fecae9_4", "scene_id": "57d9fecae9", "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": "validation"}
{"id": "57d9fecae9_5", "scene_id": "57d9fecae9", "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. laptop\nB. monitor\nC. boxes\nD. chair", "choices": ["A. laptop", "B. monitor", "C. boxes", "D. chair"], "answer": "C. boxes", "target": "notebook", "split": "validation"}
{"id": "57d9fecae9_6", "scene_id": "57d9fecae9", "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 desk. It is also to the side of and above the chair, about 17 cm away. It is further to the side of and below the monitor, about 18 cm away.", "target": "notebook", "split": "validation"}
{"id": "57d9fecae9_7", "scene_id": "57d9fecae9", "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\": \"desk\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 17.4}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 18.3}]}", "target": "notebook", "split": "validation"}
{"id": "57d9fecae9_8", "scene_id": "57d9fecae9", "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": "Yes", "target": "scissors", "split": "validation"}
{"id": "57d9fecae9_9", "scene_id": "57d9fecae9", "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. doorframe\nB. sponge\nC. pan\nD. kitchen counter", "choices": ["A. doorframe", "B. sponge", "C. pan", "D. kitchen counter"], "answer": "C. pan", "target": "scissors", "split": "validation"}
{"id": "57d9fecae9_10", "scene_id": "57d9fecae9", "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 to the side of and above the pan, about 19 cm away. It is also to the side of and above the doorframe, about 25 cm away. It is further to the side of and above the door, about 25 cm away.", "target": "scissors", "split": "validation"}
{"id": "57d9fecae9_11", "scene_id": "57d9fecae9", "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\": \"kitchen counter\", \"references\": [{\"object\": \"pan\", \"relation\": \"side_above\", \"distance_cm\": 18.6}, {\"object\": \"doorframe\", \"relation\": \"side_above\", \"distance_cm\": 25.0}]}", "target": "scissors", "split": "validation"}
{"id": "5f0fb991a7_0", "scene_id": "5f0fb991a7", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "5f0fb991a7_1", "scene_id": "5f0fb991a7", "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 desk, about 8 cm away. It is also to the side of and above the garment bag, about 8 cm away. It is further to the side of and above the luggage, about 16 cm away.", "target": "backpack", "split": "validation"}
{"id": "5f0fb991a7_2", "scene_id": "5f0fb991a7", "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\": \"desk\", \"references\": [{\"object\": \"desk\", \"relation\": \"side_below\", \"distance_cm\": 7.9}, {\"object\": \"garment bag\", \"relation\": \"side_above\", \"distance_cm\": 8.0}]}", "target": "backpack", "split": "validation"}
{"id": "5f0fb991a7_3", "scene_id": "5f0fb991a7", "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": "validation"}
{"id": "5f0fb991a7_4", "scene_id": "5f0fb991a7", "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. sink countertop\nB. light switch\nC. sink\nD. towel", "choices": ["A. sink countertop", "B. light switch", "C. sink", "D. towel"], "answer": "D. towel", "target": "toothbrush", "split": "validation"}
{"id": "5f0fb991a7_5", "scene_id": "5f0fb991a7", "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 on the shelf. It is also to the side of and above the towel, about 36 cm away. It is further to the side of and above the sink countertop, about 40 cm away.", "target": "toothbrush", "split": "validation"}
{"id": "5f0fb991a7_6", "scene_id": "5f0fb991a7", "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\": \"shelf\", \"references\": [{\"object\": \"towel\", \"relation\": \"side_above\", \"distance_cm\": 36.2}, {\"object\": \"sink countertop\", \"relation\": \"side_above\", \"distance_cm\": 40.3}]}", "target": "toothbrush", "split": "validation"}
{"id": "610eff6509_0", "scene_id": "610eff6509", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "610eff6509_1", "scene_id": "610eff6509", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. windowsill\nB. window frame\nC. chopping boards\nD. bowl", "choices": ["A. windowsill", "B. window frame", "C. chopping boards", "D. bowl"], "answer": "D. bowl", "target": "bottle", "split": "validation"}
{"id": "610eff6509_2", "scene_id": "610eff6509", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 bowl, about 6 cm away. It is also next to the windowsill, about 13 cm away. It is further to the side of and below the window frame, about 17 cm away.", "target": "bottle", "split": "validation"}
{"id": "610eff6509_3", "scene_id": "610eff6509", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"bowl\", \"relation\": \"side_above\", \"distance_cm\": 6.1}, {\"object\": \"windowsill\", \"relation\": \"next_to\", \"distance_cm\": 13.2}]}", "target": "bottle", "split": "validation"}
{"id": "612f70fe00_0", "scene_id": "612f70fe00", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "612f70fe00_1", "scene_id": "612f70fe00", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. baguettes\nB. cutting board\nC. cleaning wipes\nD. salt", "choices": ["A. baguettes", "B. cutting board", "C. cleaning wipes", "D. salt"], "answer": "C. cleaning wipes", "target": "mug", "split": "validation"}
{"id": "612f70fe00_2", "scene_id": "612f70fe00", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 cleaning wipes, about 5 cm away. It is further next to the baguettes, about 7 cm away.", "target": "mug", "split": "validation"}
{"id": "612f70fe00_3", "scene_id": "612f70fe00", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"cleaning wipes\", \"relation\": \"next_to\", \"distance_cm\": 4.8}, {\"object\": \"baguettes\", \"relation\": \"next_to\", \"distance_cm\": 6.5}]}", "target": "mug", "split": "validation"}
{"id": "612f70fe00_4", "scene_id": "612f70fe00", "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": "validation"}
{"id": "612f70fe00_5", "scene_id": "612f70fe00", "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. paper bag\nB. detergent\nC. air fryer\nD. shelf rail", "choices": ["A. paper bag", "B. detergent", "C. air fryer", "D. shelf rail"], "answer": "D. shelf rail", "target": "spray bottle", "split": "validation"}
{"id": "612f70fe00_6", "scene_id": "612f70fe00", "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 shelf rail, about 28 cm away. It is also to the side of and above the paper bag, about 73 cm away. It is further to the side of and above the detergent, about 77 cm away.", "target": "spray bottle", "split": "validation"}
{"id": "612f70fe00_7", "scene_id": "612f70fe00", "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 rail\", \"references\": [{\"object\": \"shelf rail\", \"relation\": \"side_above\", \"distance_cm\": 28.4}, {\"object\": \"paper bag\", \"relation\": \"side_above\", \"distance_cm\": 72.5}]}", "target": "spray bottle", "split": "validation"}
{"id": "612f70fe00_8", "scene_id": "612f70fe00", "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": "validation"}
{"id": "612f70fe00_9", "scene_id": "612f70fe00", "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. table\nB. cutting board\nC. baguettes\nD. knife", "choices": ["A. table", "B. cutting board", "C. baguettes", "D. knife"], "answer": "A. table", "target": "tote bag", "split": "validation"}
{"id": "612f70fe00_10", "scene_id": "612f70fe00", "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 under the table, about 4 cm away. It is also to the side of and below the cutting board, about 6 cm away. It is further to the side of and below the knife, about 11 cm away.", "target": "tote bag", "split": "validation"}
{"id": "612f70fe00_11", "scene_id": "612f70fe00", "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\": \"table\", \"relation\": \"under\", \"distance_cm\": 4.5}, {\"object\": \"cutting board\", \"relation\": \"side_below\", \"distance_cm\": 6.3}]}", "target": "tote bag", "split": "validation"}
{"id": "618310ed87_0", "scene_id": "618310ed87", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "618310ed87_1", "scene_id": "618310ed87", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. pot\nC. picture\nD. light switch", "choices": ["A. power socket", "B. pot", "C. picture", "D. light switch"], "answer": "B. pot", "target": "backpack", "split": "validation"}
{"id": "618310ed87_2", "scene_id": "618310ed87", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 pot, about 28 cm away. It is also to the side of and below the picture, about 33 cm away. It is further to the side of and below the power socket, about 36 cm away.", "target": "backpack", "split": "validation"}
{"id": "618310ed87_3", "scene_id": "618310ed87", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"pot\", \"references\": [{\"object\": \"pot\", \"relation\": \"side_below\", \"distance_cm\": 28.4}, {\"object\": \"picture\", \"relation\": \"side_below\", \"distance_cm\": 33.1}]}", "target": "backpack", "split": "validation"}
{"id": "618310ed87_4", "scene_id": "618310ed87", "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": "validation"}
{"id": "618310ed87_5", "scene_id": "618310ed87", "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. dish sponge\nB. dish cloth\nC. cloth\nD. heater", "choices": ["A. dish sponge", "B. dish cloth", "C. cloth", "D. heater"], "answer": "C. cloth", "target": "bottle", "split": "validation"}
{"id": "618310ed87_6", "scene_id": "618310ed87", "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 to the side of and above the cloth, about 5 cm away. It is further to the side of and above the heater, about 34 cm away.", "target": "bottle", "split": "validation"}
{"id": "618310ed87_7", "scene_id": "618310ed87", "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\": \"cloth\", \"relation\": \"side_above\", \"distance_cm\": 4.6}, {\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 33.6}]}", "target": "bottle", "split": "validation"}
{"id": "618310ed87_8", "scene_id": "618310ed87", "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": "validation"}
{"id": "618310ed87_9", "scene_id": "618310ed87", "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. power sockets\nB. table\nC. chair\nD. power socket", "choices": ["A. power sockets", "B. table", "C. chair", "D. power socket"], "answer": "B. table", "target": "handbag", "split": "validation"}
{"id": "618310ed87_10", "scene_id": "618310ed87", "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 table, about 6 cm away. It is also to the side of and below the power sockets, about 64 cm away. It is further near the power socket, about 71 cm away.", "target": "handbag", "split": "validation"}
{"id": "618310ed87_11", "scene_id": "618310ed87", "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\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 6.4}, {\"object\": \"power sockets\", \"relation\": \"side_below\", \"distance_cm\": 64.2}]}", "target": "handbag", "split": "validation"}
{"id": "618310ed87_12", "scene_id": "618310ed87", "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 smartphone appear?", "choices": null, "answer": "No", "target": "smartphone", "split": "validation"}
{"id": "618310ed87_13", "scene_id": "618310ed87", "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 to the side of and below the faucet, about 31 cm away. It is also near the dish sponge, about 31 cm away. It is further to the side of and above the dish cloth, about 50 cm away.", "target": "mug", "split": "validation"}
{"id": "618310ed87_14", "scene_id": "618310ed87", "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\": \"sink\", \"references\": [{\"object\": \"faucet\", \"relation\": \"side_below\", \"distance_cm\": 30.6}, {\"object\": \"dish sponge\", \"relation\": \"near\", \"distance_cm\": 31.0}]}", "target": "mug", "split": "validation"}
{"id": "618310ed87_15", "scene_id": "618310ed87", "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 tv remote appear?", "choices": null, "answer": "Yes", "target": "tv remote", "split": "validation"}
{"id": "618310ed87_16", "scene_id": "618310ed87", "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 tv remote in this scene?\nA. pot\nB. picture\nC. tv\nD. backpack", "choices": ["A. pot", "B. picture", "C. tv", "D. backpack"], "answer": "C. tv", "target": "tv remote", "split": "validation"}
{"id": "618310ed87_17", "scene_id": "618310ed87", "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 tv remote in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tv remote is under the tv, about 3 cm away. It is also to the side of and below the picture, about 8 cm away. It is further to the side of and below the pot, about 51 cm away.", "target": "tv remote", "split": "validation"}
{"id": "618310ed87_18", "scene_id": "618310ed87", "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 tv remote in structured JSON format.", "choices": null, "answer": "{\"target\": \"tv remote\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"tv\", \"relation\": \"under\", \"distance_cm\": 2.6}, {\"object\": \"picture\", \"relation\": \"side_below\", \"distance_cm\": 7.8}]}", "target": "tv remote", "split": "validation"}
{"id": "639f2c4d5a_0", "scene_id": "639f2c4d5a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "639f2c4d5a_1", "scene_id": "639f2c4d5a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. faucet\nB. makeup pen\nC. soap dispenser\nD. shelf", "choices": ["A. faucet", "B. makeup pen", "C. soap dispenser", "D. shelf"], "answer": "B. makeup pen", "target": "cup", "split": "validation"}
{"id": "639f2c4d5a_2", "scene_id": "639f2c4d5a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 above the makeup pen. It is also next to the faucet, about 6 cm away. It is further to the side of and below the shelf, about 10 cm away.", "target": "cup", "split": "validation"}
{"id": "639f2c4d5a_3", "scene_id": "639f2c4d5a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"makeup pen\", \"relation\": \"above\", \"distance_cm\": 2.3}, {\"object\": \"faucet\", \"relation\": \"next_to\", \"distance_cm\": 6.2}]}", "target": "cup", "split": "validation"}
{"id": "64b8886d68_0", "scene_id": "64b8886d68", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "64b8886d68_1", "scene_id": "64b8886d68", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. light switch\nC. monitor\nD. power socket", "choices": ["A. table", "B. light switch", "C. monitor", "D. power socket"], "answer": "B. light switch", "target": "phone charger", "split": "validation"}
{"id": "64b8886d68_2", "scene_id": "64b8886d68", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 76 cm away.", "target": "phone charger", "split": "validation"}
{"id": "64b8886d68_3", "scene_id": "64b8886d68", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 10.0}, {\"object\": \"light switch\", \"relation\": \"below\", \"distance_cm\": 76.1}]}", "target": "phone charger", "split": "validation"}
{"id": "666d04a14a_0", "scene_id": "666d04a14a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "666d04a14a_1", "scene_id": "666d04a14a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. doorframe\nB. shelf\nC. box\nD. apple", "choices": ["A. doorframe", "B. shelf", "C. box", "D. apple"], "answer": "A. doorframe", "target": "backpack", "split": "validation"}
{"id": "666d04a14a_2", "scene_id": "666d04a14a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 doorframe, about 5 cm away. It is also to the side of and above the door, about 5 cm away. It is further to the side of and below the shelf, about 24 cm away.", "target": "backpack", "split": "validation"}
{"id": "666d04a14a_3", "scene_id": "666d04a14a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"doorframe\", \"relation\": \"side_above\", \"distance_cm\": 4.7}, {\"object\": \"door\", \"relation\": \"side_above\", \"distance_cm\": 5.3}]}", "target": "backpack", "split": "validation"}
{"id": "666d04a14a_4", "scene_id": "666d04a14a", "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 sticky note appear?", "choices": null, "answer": "No", "target": "sticky note", "split": "validation"}
{"id": "666d04a14a_5", "scene_id": "666d04a14a", "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. towel\nB. cloth\nC. book\nD. tray", "choices": ["A. towel", "B. cloth", "C. book", "D. tray"], "answer": "D. tray", "target": "bottle", "split": "validation"}
{"id": "666d04a14a_6", "scene_id": "666d04a14a", "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 tray, about 3 cm away. It is further next to the cloth, about 9 cm away.", "target": "bottle", "split": "validation"}
{"id": "666d04a14a_7", "scene_id": "666d04a14a", "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\": \"tray\", \"relation\": \"next_to\", \"distance_cm\": 3.1}, {\"object\": \"cloth\", \"relation\": \"next_to\", \"distance_cm\": 8.6}]}", "target": "bottle", "split": "validation"}
{"id": "666d04a14a_8", "scene_id": "666d04a14a", "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": "validation"}
{"id": "666d04a14a_9", "scene_id": "666d04a14a", "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. document tray\nB. cabinet\nC. cloth\nD. desk orginizer", "choices": ["A. document tray", "B. cabinet", "C. cloth", "D. desk orginizer"], "answer": "B. cabinet", "target": "power strip", "split": "validation"}
{"id": "666d04a14a_10", "scene_id": "666d04a14a", "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 cabinet, about 12 cm away. It is further to the side of and below the document tray, about 13 cm away.", "target": "power strip", "split": "validation"}
{"id": "666d04a14a_11", "scene_id": "666d04a14a", "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\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 11.6}, {\"object\": \"document tray\", \"relation\": \"side_below\", \"distance_cm\": 13.2}]}", "target": "power strip", "split": "validation"}
{"id": "67ab69f888_0", "scene_id": "67ab69f888", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "67ab69f888_1", "scene_id": "67ab69f888", "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. It is also next to the mug. It is further to the side of and below the monitor, about 7 cm away.", "target": "bottle", "split": "validation"}
{"id": "67ab69f888_2", "scene_id": "67ab69f888", "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\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 2.1}, {\"object\": \"mug\", \"relation\": \"next_to\", \"distance_cm\": 2.3}]}", "target": "bottle", "split": "validation"}
{"id": "67ab69f888_3", "scene_id": "67ab69f888", "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": "No", "target": "tv remote", "split": "validation"}
{"id": "67ab69f888_4", "scene_id": "67ab69f888", "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 on the table. It is also next to the mousepad, about 3 cm away. It is further next to the mouse, about 4 cm away.", "target": "keyboard", "split": "validation"}
{"id": "67ab69f888_5", "scene_id": "67ab69f888", "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\": \"table\", \"references\": [{\"object\": \"mousepad\", \"relation\": \"next_to\", \"distance_cm\": 3.4}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 3.9}]}", "target": "keyboard", "split": "validation"}
{"id": "67ab69f888_6", "scene_id": "67ab69f888", "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 adapter appear?", "choices": null, "answer": "No", "target": "adapter", "split": "validation"}
{"id": "67ab69f888_7", "scene_id": "67ab69f888", "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 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 4 cm away. It is further to the side of and below the monitor, about 20 cm away.", "target": "mouse", "split": "validation"}
{"id": "67ab69f888_8", "scene_id": "67ab69f888", "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 mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 3.9}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 4.0}]}", "target": "mouse", "split": "validation"}
{"id": "67ab69f888_9", "scene_id": "67ab69f888", "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 remote controller appear?", "choices": null, "answer": "No", "target": "remote controller", "split": "validation"}
{"id": "67ab69f888_10", "scene_id": "67ab69f888", "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 mousepad in this scene?\nA. keyboard\nB. mug\nC. monitor\nD. bottle", "choices": ["A. keyboard", "B. mug", "C. monitor", "D. bottle"], "answer": "A. keyboard", "target": "mousepad", "split": "validation"}
{"id": "67ab69f888_11", "scene_id": "67ab69f888", "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 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 monitor, about 8 cm away.", "target": "mousepad", "split": "validation"}
{"id": "67ab69f888_12", "scene_id": "67ab69f888", "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 mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 3.4}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 7.6}]}", "target": "mousepad", "split": "validation"}
{"id": "67ab69f888_13", "scene_id": "67ab69f888", "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 mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "validation"}
{"id": "67ab69f888_14", "scene_id": "67ab69f888", "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 mug in this scene?\nA. telephone\nB. bottle\nC. mousepad\nD. monitor", "choices": ["A. telephone", "B. bottle", "C. mousepad", "D. monitor"], "answer": "B. bottle", "target": "mug", "split": "validation"}
{"id": "67ab69f888_15", "scene_id": "67ab69f888", "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 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. It is further to the side of and below the monitor, about 6 cm away.", "target": "mug", "split": "validation"}
{"id": "67ab69f888_16", "scene_id": "67ab69f888", "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 mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"bottle\", \"relation\": \"next_to\", \"distance_cm\": 2.3}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 6.5}]}", "target": "mug", "split": "validation"}
{"id": "67ab69f888_17", "scene_id": "67ab69f888", "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 pen appear?", "choices": null, "answer": "Yes", "target": "pen", "split": "validation"}
{"id": "67ab69f888_18", "scene_id": "67ab69f888", "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 pen in this scene?\nA. bottle\nB. computer tower\nC. trash bin\nD. file binder", "choices": ["A. bottle", "B. computer tower", "C. trash bin", "D. file binder"], "answer": "D. file binder", "target": "pen", "split": "validation"}
{"id": "67ab69f888_19", "scene_id": "67ab69f888", "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 pen in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pen is to the side of and below the file binder, about 10 cm away. It is also to the side of and above the trash bin, about 74 cm away.", "target": "pen", "split": "validation"}
{"id": "67ab69f888_20", "scene_id": "67ab69f888", "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 pen in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen\", \"support_surface\": \"file rack\", \"references\": [{\"object\": \"file binder\", \"relation\": \"side_below\", \"distance_cm\": 9.6}, {\"object\": \"trash bin\", \"relation\": \"side_above\", \"distance_cm\": 73.7}]}", "target": "pen", "split": "validation"}
{"id": "6855e1ac32_0", "scene_id": "6855e1ac32", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "6855e1ac32_1", "scene_id": "6855e1ac32", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. shelf\nB. table\nC. kettle\nD. cup", "choices": ["A. shelf", "B. table", "C. kettle", "D. cup"], "answer": "B. table", "target": "backpack", "split": "validation"}
{"id": "6855e1ac32_2", "scene_id": "6855e1ac32", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 15 cm away. It is also to the side of and below the shelf, about 32 cm away. It is further to the side of and below the kettle, about 43 cm away.", "target": "backpack", "split": "validation"}
{"id": "6855e1ac32_3", "scene_id": "6855e1ac32", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 15.1}, {\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 32.1}]}", "target": "backpack", "split": "validation"}
{"id": "6855e1ac32_4", "scene_id": "6855e1ac32", "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": "validation"}
{"id": "6855e1ac32_5", "scene_id": "6855e1ac32", "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. sink\nB. tap\nC. split\nD. towel", "choices": ["A. sink", "B. tap", "C. split", "D. towel"], "answer": "B. tap", "target": "bottle", "split": "validation"}
{"id": "6855e1ac32_6", "scene_id": "6855e1ac32", "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 to the side of and above the tap, about 8 cm away. It is further to the side of and above the sink, about 13 cm away.", "target": "bottle", "split": "validation"}
{"id": "6855e1ac32_7", "scene_id": "6855e1ac32", "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\": \"tap\", \"relation\": \"side_above\", \"distance_cm\": 8.2}, {\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 12.8}]}", "target": "bottle", "split": "validation"}
{"id": "6855e1ac32_8", "scene_id": "6855e1ac32", "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": "validation"}
{"id": "6855e1ac32_9", "scene_id": "6855e1ac32", "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 purse in this scene?\nA. cabinet\nB. ottoman\nC. light switch\nD. jacket", "choices": ["A. cabinet", "B. ottoman", "C. light switch", "D. jacket"], "answer": "A. cabinet", "target": "purse", "split": "validation"}
{"id": "6855e1ac32_10", "scene_id": "6855e1ac32", "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 purse in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The purse is on the shelf. It is also to the side of and below the cabinet, about 9 cm away. It is further to the side of and below the jacket, about 35 cm away.", "target": "purse", "split": "validation"}
{"id": "6855e1ac32_11", "scene_id": "6855e1ac32", "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 purse in structured JSON format.", "choices": null, "answer": "{\"target\": \"purse\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 8.5}, {\"object\": \"jacket\", \"relation\": \"side_below\", \"distance_cm\": 34.7}]}", "target": "purse", "split": "validation"}
{"id": "689fec23d7_0", "scene_id": "689fec23d7", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "689fec23d7_1", "scene_id": "689fec23d7", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. office chair\nB. phone\nC. cup\nD. keyboard", "choices": ["A. office chair", "B. phone", "C. cup", "D. keyboard"], "answer": "B. phone", "target": "headset", "split": "validation"}
{"id": "689fec23d7_2", "scene_id": "689fec23d7", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 desk. It is also next to the phone. It is further next to the cup, about 5 cm away.", "target": "headset", "split": "validation"}
{"id": "689fec23d7_3", "scene_id": "689fec23d7", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"desk\", \"references\": [{\"object\": \"phone\", \"relation\": \"next_to\", \"distance_cm\": 2.5}, {\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 4.9}]}", "target": "headset", "split": "validation"}
{"id": "689fec23d7_4", "scene_id": "689fec23d7", "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": "validation"}
{"id": "689fec23d7_5", "scene_id": "689fec23d7", "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. cup\nC. office chair\nD. monitor", "choices": ["A. keyboard", "B. cup", "C. office chair", "D. monitor"], "answer": "B. cup", "target": "mouse", "split": "validation"}
{"id": "689fec23d7_6", "scene_id": "689fec23d7", "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 cup, about 6 cm away. It is further next to the keyboard, about 7 cm away.", "target": "mouse", "split": "validation"}
{"id": "689fec23d7_7", "scene_id": "689fec23d7", "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\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 5.5}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 7.3}]}", "target": "mouse", "split": "validation"}
{"id": "6c75a5290e_0", "scene_id": "6c75a5290e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "6c75a5290e_1", "scene_id": "6c75a5290e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. blanket\nC. books\nD. bedside cabinet", "choices": ["A. backpack", "B. blanket", "C. books", "D. bedside cabinet"], "answer": "B. blanket", "target": "bottle", "split": "validation"}
{"id": "6c75a5290e_2", "scene_id": "6c75a5290e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 blanket, about 33 cm away.", "target": "bottle", "split": "validation"}
{"id": "6c75a5290e_3", "scene_id": "6c75a5290e", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"blanket\", \"relation\": \"side_below\", \"distance_cm\": 32.6}]}", "target": "bottle", "split": "validation"}
{"id": "6c75a5290e_4", "scene_id": "6c75a5290e", "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": "No", "target": "remote control", "split": "validation"}
{"id": "6c75a5290e_5", "scene_id": "6c75a5290e", "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. tablet\nB. books\nC. phone\nD. table lamp", "choices": ["A. tablet", "B. books", "C. phone", "D. table lamp"], "answer": "A. tablet", "target": "pen holder", "split": "validation"}
{"id": "6c75a5290e_6", "scene_id": "6c75a5290e", "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 on the cabinet. It is also to the side of and above the tablet, about 4 cm away. It is further next to the books, about 9 cm away.", "target": "pen holder", "split": "validation"}
{"id": "6c75a5290e_7", "scene_id": "6c75a5290e", "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\": \"cabinet\", \"references\": [{\"object\": \"tablet\", \"relation\": \"side_above\", \"distance_cm\": 4.0}, {\"object\": \"books\", \"relation\": \"next_to\", \"distance_cm\": 9.4}]}", "target": "pen holder", "split": "validation"}
{"id": "6c75a5290e_8", "scene_id": "6c75a5290e", "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 appear?", "choices": null, "answer": "Yes", "target": "phone", "split": "validation"}
{"id": "6c75a5290e_9", "scene_id": "6c75a5290e", "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 in this scene?\nA. table lamp\nB. tablet\nC. books\nD. pen holder", "choices": ["A. table lamp", "B. tablet", "C. books", "D. pen holder"], "answer": "B. tablet", "target": "phone", "split": "validation"}
{"id": "6c75a5290e_10", "scene_id": "6c75a5290e", "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 in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone is on the cabinet. It is also next to the tablet. It is further to the side of and below the table lamp, about 12 cm away.", "target": "phone", "split": "validation"}
{"id": "6c75a5290e_11", "scene_id": "6c75a5290e", "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 in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"tablet\", \"relation\": \"next_to\", \"distance_cm\": 2.0}, {\"object\": \"table lamp\", \"relation\": \"side_below\", \"distance_cm\": 11.9}]}", "target": "phone", "split": "validation"}
{"id": "6cc2231b9c_0", "scene_id": "6cc2231b9c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "6cc2231b9c_1", "scene_id": "6cc2231b9c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. table\nB. keyboard\nC. monitor\nD. telephone", "choices": ["A. table", "B. keyboard", "C. monitor", "D. telephone"], "answer": "A. table", "target": "pen holder", "split": "validation"}
{"id": "6cc2231b9c_2", "scene_id": "6cc2231b9c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 monitor, about 12 cm away. It is further near the keyboard, about 40 cm away.", "target": "pen holder", "split": "validation"}
{"id": "6cc2231b9c_3", "scene_id": "6cc2231b9c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"side_below\", \"distance_cm\": 11.6}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 39.8}]}", "target": "pen holder", "split": "validation"}
{"id": "6e420fd87a_0", "scene_id": "6e420fd87a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "6e420fd87a_1", "scene_id": "6e420fd87a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. outlet\nC. chair\nD. pillar", "choices": ["A. power socket", "B. outlet", "C. chair", "D. pillar"], "answer": "D. pillar", "target": "power strip", "split": "validation"}
{"id": "6e420fd87a_2", "scene_id": "6e420fd87a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 pillar, about 27 cm away. It is also to the side of and above the chair, about 46 cm away. It is further near the power socket, about 72 cm away.", "target": "power strip", "split": "validation"}
{"id": "6e420fd87a_3", "scene_id": "6e420fd87a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"classroom table\", \"references\": [{\"object\": \"pillar\", \"relation\": \"side_below\", \"distance_cm\": 26.6}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 46.5}]}", "target": "power strip", "split": "validation"}
{"id": "6ef2ac745a_0", "scene_id": "6ef2ac745a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "6ef2ac745a_1", "scene_id": "6ef2ac745a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. computer tower\nC. flipchart\nD. chair", "choices": ["A. power socket", "B. computer tower", "C. flipchart", "D. chair"], "answer": "A. power socket", "target": "backpack", "split": "validation"}
{"id": "6ef2ac745a_2", "scene_id": "6ef2ac745a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 79 cm away.", "target": "backpack", "split": "validation"}
{"id": "6ef2ac745a_3", "scene_id": "6ef2ac745a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"flipchart\", \"references\": [{\"object\": \"power socket\", \"relation\": \"side_below\", \"distance_cm\": 79.1}]}", "target": "backpack", "split": "validation"}
{"id": "6ef2ac745a_4", "scene_id": "6ef2ac745a", "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": "validation"}
{"id": "6ef2ac745a_5", "scene_id": "6ef2ac745a", "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. computer tower\nB. monitor\nC. chair\nD. office partition", "choices": ["A. computer tower", "B. monitor", "C. chair", "D. office partition"], "answer": "B. monitor", "target": "keyboard", "split": "validation"}
{"id": "6ef2ac745a_6", "scene_id": "6ef2ac745a", "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 11 cm away. It is further to the side of and above the chair, about 32 cm away.", "target": "keyboard", "split": "validation"}
{"id": "6ef2ac745a_7", "scene_id": "6ef2ac745a", "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\": 10.9}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 32.3}]}", "target": "keyboard", "split": "validation"}
{"id": "70945f435a_0", "scene_id": "70945f435a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "70945f435a_1", "scene_id": "70945f435a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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 kettle\nB. refrigerator\nC. tray\nD. magnet", "choices": ["A. electric kettle", "B. refrigerator", "C. tray", "D. magnet"], "answer": "C. tray", "target": "bottle", "split": "validation"}
{"id": "70945f435a_2", "scene_id": "70945f435a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 tray, about 3 cm away. It is also to the side of and above the refrigerator, about 19 cm away. It is further to the side of and below the electric kettle, about 23 cm away.", "target": "bottle", "split": "validation"}
{"id": "70945f435a_3", "scene_id": "70945f435a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"tray\", \"relation\": \"under\", \"distance_cm\": 3.0}, {\"object\": \"refrigerator\", \"relation\": \"side_above\", \"distance_cm\": 19.0}]}", "target": "bottle", "split": "validation"}
{"id": "73f9370962_0", "scene_id": "73f9370962", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "73f9370962_1", "scene_id": "73f9370962", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. windowsill\nD. paper", "choices": ["A. monitor", "B. mouse", "C. windowsill", "D. paper"], "answer": "A. monitor", "target": "keyboard", "split": "validation"}
{"id": "73f9370962_2", "scene_id": "73f9370962", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 cabinet. It is also to the side of and below the monitor, about 12 cm away. It is further to the side of and below the paper, about 16 cm away.", "target": "keyboard", "split": "validation"}
{"id": "73f9370962_3", "scene_id": "73f9370962", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\", \"relation\": \"side_below\", \"distance_cm\": 12.4}, {\"object\": \"paper\", \"relation\": \"side_below\", \"distance_cm\": 15.8}]}", "target": "keyboard", "split": "validation"}
{"id": "73f9370962_4", "scene_id": "73f9370962", "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": "validation"}
{"id": "73f9370962_5", "scene_id": "73f9370962", "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. windowsill\nC. monitor\nD. keyboard", "choices": ["A. paper", "B. windowsill", "C. monitor", "D. keyboard"], "answer": "C. monitor", "target": "mouse", "split": "validation"}
{"id": "73f9370962_6", "scene_id": "73f9370962", "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 cabinet. 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 paper, about 29 cm away.", "target": "mouse", "split": "validation"}
{"id": "73f9370962_7", "scene_id": "73f9370962", "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\": \"cabinet\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 26.2}, {\"object\": \"paper\", \"relation\": \"side_below\", \"distance_cm\": 29.2}]}", "target": "mouse", "split": "validation"}
{"id": "778769dad8_0", "scene_id": "778769dad8", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "778769dad8_1", "scene_id": "778769dad8", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. paper bag\nC. laptop\nD. desk", "choices": ["A. books", "B. paper bag", "C. laptop", "D. desk"], "answer": "D. desk", "target": "backpack", "split": "validation"}
{"id": "778769dad8_2", "scene_id": "778769dad8", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 desk, about 16 cm away. It is also near the laptop, about 43 cm away. It is further to the side of and below the books, about 58 cm away.", "target": "backpack", "split": "validation"}
{"id": "778769dad8_3", "scene_id": "778769dad8", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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_above\", \"distance_cm\": 16.1}, {\"object\": \"laptop\", \"relation\": \"near\", \"distance_cm\": 43.2}]}", "target": "backpack", "split": "validation"}
{"id": "778769dad8_4", "scene_id": "778769dad8", "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": "validation"}
{"id": "778769dad8_5", "scene_id": "778769dad8", "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. kitchen utensils\nB. basket\nC. kitchen appliance\nD. jar", "choices": ["A. kitchen utensils", "B. basket", "C. kitchen appliance", "D. jar"], "answer": "D. jar", "target": "cup", "split": "validation"}
{"id": "778769dad8_6", "scene_id": "778769dad8", "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 jar, about 5 cm away. It is also to the side of and above the kitchen appliance, about 24 cm away. It is further to the side of and below the basket, about 25 cm away.", "target": "cup", "split": "validation"}
{"id": "778769dad8_7", "scene_id": "778769dad8", "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\": \"jar\", \"relation\": \"next_to\", \"distance_cm\": 4.9}, {\"object\": \"kitchen appliance\", \"relation\": \"side_above\", \"distance_cm\": 23.9}]}", "target": "cup", "split": "validation"}
{"id": "778769dad8_8", "scene_id": "778769dad8", "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": "validation"}
{"id": "778769dad8_9", "scene_id": "778769dad8", "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 next to the books, about 5 cm away. It is further under the bottle, about 5 cm away.", "target": "headphones", "split": "validation"}
{"id": "778769dad8_10", "scene_id": "778769dad8", "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\": \"books\", \"relation\": \"next_to\", \"distance_cm\": 4.7}, {\"object\": \"bottle\", \"relation\": \"under\", \"distance_cm\": 4.8}]}", "target": "headphones", "split": "validation"}
{"id": "778769dad8_11", "scene_id": "778769dad8", "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 headset appear?", "choices": null, "answer": "No", "target": "headset", "split": "validation"}
{"id": "778769dad8_12", "scene_id": "778769dad8", "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. chair\nB. books\nC. headphones\nD. laptop", "choices": ["A. chair", "B. books", "C. headphones", "D. laptop"], "answer": "D. laptop", "target": "phone stand", "split": "validation"}
{"id": "778769dad8_13", "scene_id": "778769dad8", "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 laptop, about 13 cm away. It is further near the books, about 17 cm away.", "target": "phone stand", "split": "validation"}
{"id": "778769dad8_14", "scene_id": "778769dad8", "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\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 13.2}, {\"object\": \"books\", \"relation\": \"near\", \"distance_cm\": 17.4}]}", "target": "phone stand", "split": "validation"}
{"id": "778769dad8_15", "scene_id": "778769dad8", "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": "validation"}
{"id": "778769dad8_16", "scene_id": "778769dad8", "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. pillow\nB. mattress\nC. heater\nD. electric kettle", "choices": ["A. pillow", "B. mattress", "C. heater", "D. electric kettle"], "answer": "D. electric kettle", "target": "power strip", "split": "validation"}
{"id": "778769dad8_17", "scene_id": "778769dad8", "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 next to the electric kettle, about 7 cm away. It is also to the side of and below the mattress, about 24 cm away. It is further to the side of and below the heater, about 31 cm away.", "target": "power strip", "split": "validation"}
{"id": "778769dad8_18", "scene_id": "778769dad8", "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\": \"mattress\", \"references\": [{\"object\": \"electric kettle\", \"relation\": \"next_to\", \"distance_cm\": 7.0}, {\"object\": \"mattress\", \"relation\": \"side_below\", \"distance_cm\": 23.6}]}", "target": "power strip", "split": "validation"}
{"id": "7968e15e5e_0", "scene_id": "7968e15e5e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "7968e15e5e_1", "scene_id": "7968e15e5e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. door frame\nB. storage cabinet\nC. backpack\nD. book", "choices": ["A. door frame", "B. storage cabinet", "C. backpack", "D. book"], "answer": "B. storage cabinet", "target": "marker", "split": "validation"}
{"id": "7968e15e5e_2", "scene_id": "7968e15e5e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 near the storage cabinet, about 66 cm away. It is also to the side of and below the door, about 71 cm away. It is further to the side of and below the door frame, about 76 cm away.", "target": "marker", "split": "validation"}
{"id": "7968e15e5e_3", "scene_id": "7968e15e5e", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"storage cabinet\", \"references\": [{\"object\": \"storage cabinet\", \"relation\": \"near\", \"distance_cm\": 65.6}, {\"object\": \"door\", \"relation\": \"side_below\", \"distance_cm\": 70.9}]}", "target": "marker", "split": "validation"}
{"id": "7968e15e5e_4", "scene_id": "7968e15e5e", "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": "validation"}
{"id": "7968e15e5e_5", "scene_id": "7968e15e5e", "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 pad in this scene?\nA. sticky note\nB. usb hub\nC. file cabinet\nD. monitor", "choices": ["A. sticky note", "B. usb hub", "C. file cabinet", "D. monitor"], "answer": "A. sticky note", "target": "mouse pad", "split": "validation"}
{"id": "7968e15e5e_6", "scene_id": "7968e15e5e", "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 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 sticky note, about 12 cm away. It is further to the side of and below the monitor, about 19 cm away.", "target": "mouse pad", "split": "validation"}
{"id": "7968e15e5e_7", "scene_id": "7968e15e5e", "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 pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"sticky note\", \"relation\": \"next_to\", \"distance_cm\": 11.7}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 19.3}]}", "target": "mouse pad", "split": "validation"}
{"id": "7968e15e5e_8", "scene_id": "7968e15e5e", "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": "Yes", "target": "usb hub", "split": "validation"}
{"id": "7968e15e5e_9", "scene_id": "7968e15e5e", "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 usb hub in this scene?\nA. hand sanitizer\nB. sticky note\nC. mouse pad\nD. laptop", "choices": ["A. hand sanitizer", "B. sticky note", "C. mouse pad", "D. laptop"], "answer": "B. sticky note", "target": "usb hub", "split": "validation"}
{"id": "7968e15e5e_10", "scene_id": "7968e15e5e", "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 usb hub in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The usb hub is on the table. It is also next to the sticky note, about 5 cm away. It is further to the side of and below the laptop, about 7 cm away.", "target": "usb hub", "split": "validation"}
{"id": "7968e15e5e_11", "scene_id": "7968e15e5e", "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 usb hub in structured JSON format.", "choices": null, "answer": "{\"target\": \"usb hub\", \"support_surface\": \"table\", \"references\": [{\"object\": \"sticky note\", \"relation\": \"next_to\", \"distance_cm\": 5.3}, {\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 6.6}]}", "target": "usb hub", "split": "validation"}
{"id": "7a629f789c_0", "scene_id": "7a629f789c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "7a629f789c_1", "scene_id": "7a629f789c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. pillow\nC. shelf\nD. cup", "choices": ["A. trash can", "B. pillow", "C. shelf", "D. cup"], "answer": "A. trash can", "target": "bottle", "split": "validation"}
{"id": "7a629f789c_2", "scene_id": "7a629f789c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 trash can, about 19 cm away. It is further near the cup, about 31 cm away.", "target": "bottle", "split": "validation"}
{"id": "7a629f789c_3", "scene_id": "7a629f789c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"trash can\", \"relation\": \"side_above\", \"distance_cm\": 18.8}, {\"object\": \"cup\", \"relation\": \"near\", \"distance_cm\": 30.7}]}", "target": "bottle", "split": "validation"}
{"id": "7b04052ad0_0", "scene_id": "7b04052ad0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "7b04052ad0_1", "scene_id": "7b04052ad0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. power strip\nC. dustbin\nD. table", "choices": ["A. box", "B. power strip", "C. dustbin", "D. table"], "answer": "A. box", "target": "backpack", "split": "validation"}
{"id": "7b04052ad0_2", "scene_id": "7b04052ad0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 rug, about 3 cm away. It is also to the side of and above the power strip, about 16 cm away. It is further to the side of and below the table, about 19 cm away.", "target": "backpack", "split": "validation"}
{"id": "7b04052ad0_3", "scene_id": "7b04052ad0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"rug\", \"relation\": \"side_above\", \"distance_cm\": 2.6}, {\"object\": \"power strip\", \"relation\": \"side_above\", \"distance_cm\": 16.0}]}", "target": "backpack", "split": "validation"}
{"id": "7b04052ad0_4", "scene_id": "7b04052ad0", "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": "validation"}
{"id": "7b04052ad0_5", "scene_id": "7b04052ad0", "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. box\nC. paper\nD. cabinet", "choices": ["A. bottle", "B. box", "C. paper", "D. cabinet"], "answer": "C. paper", "target": "cup", "split": "validation"}
{"id": "7b04052ad0_6", "scene_id": "7b04052ad0", "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 paper, about 7 cm away. It is also to the side of and above the box, about 27 cm away. It is further to the side of and below the cabinet, about 29 cm away.", "target": "cup", "split": "validation"}
{"id": "7b04052ad0_7", "scene_id": "7b04052ad0", "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\": \"side table\", \"references\": [{\"object\": \"paper\", \"relation\": \"side_below\", \"distance_cm\": 6.8}, {\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 26.9}]}", "target": "cup", "split": "validation"}
{"id": "7b04052ad0_8", "scene_id": "7b04052ad0", "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": "validation"}
{"id": "7b04052ad0_9", "scene_id": "7b04052ad0", "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. chair\nB. laptop\nC. cloth\nD. toilet paper roll", "choices": ["A. chair", "B. laptop", "C. cloth", "D. toilet paper roll"], "answer": "B. laptop", "target": "keyboard", "split": "validation"}
{"id": "7b04052ad0_10", "scene_id": "7b04052ad0", "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 laptop, about 12 cm away. It is further to the side of and below the cloth, about 14 cm away.", "target": "keyboard", "split": "validation"}
{"id": "7b04052ad0_11", "scene_id": "7b04052ad0", "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\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 12.0}, {\"object\": \"cloth\", \"relation\": \"side_below\", \"distance_cm\": 13.8}]}", "target": "keyboard", "split": "validation"}
{"id": "7b04052ad0_12", "scene_id": "7b04052ad0", "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": "validation"}
{"id": "7b04052ad0_13", "scene_id": "7b04052ad0", "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. satchel\nB. box\nC. backpack\nD. heater", "choices": ["A. satchel", "B. box", "C. backpack", "D. heater"], "answer": "D. heater", "target": "power strip", "split": "validation"}
{"id": "7b04052ad0_14", "scene_id": "7b04052ad0", "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 rug. It is also to the side of and below the heater, about 10 cm away. It is further to the side of and below the backpack, about 16 cm away.", "target": "power strip", "split": "validation"}
{"id": "7b04052ad0_15", "scene_id": "7b04052ad0", "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\": \"rug\", \"references\": [{\"object\": \"heater\", \"relation\": \"side_below\", \"distance_cm\": 9.6}, {\"object\": \"backpack\", \"relation\": \"side_below\", \"distance_cm\": 16.0}]}", "target": "power strip", "split": "validation"}
{"id": "7b04052ad0_16", "scene_id": "7b04052ad0", "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 tape dispenser appear?", "choices": null, "answer": "No", "target": "tape dispenser", "split": "validation"}
{"id": "7b04052ad0_17", "scene_id": "7b04052ad0", "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 soap in this scene?\nA. air fryer\nB. window sill\nC. dish sponge\nD. bottle", "choices": ["A. air fryer", "B. window sill", "C. dish sponge", "D. bottle"], "answer": "D. bottle", "target": "soap", "split": "validation"}
{"id": "7b04052ad0_18", "scene_id": "7b04052ad0", "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 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 bottle, about 43 cm away. It is also to the side of and below the window sill, about 58 cm away. It is further to the side of and below the air fryer, about 70 cm away.", "target": "soap", "split": "validation"}
{"id": "7b04052ad0_19", "scene_id": "7b04052ad0", "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 soap in structured JSON format.", "choices": null, "answer": "{\"target\": \"soap\", \"support_surface\": \"window sill\", \"references\": [{\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 43.4}, {\"object\": \"window sill\", \"relation\": \"side_below\", \"distance_cm\": 58.3}]}", "target": "soap", "split": "validation"}
{"id": "7bc286c1b6_0", "scene_id": "7bc286c1b6", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "7bc286c1b6_1", "scene_id": "7bc286c1b6", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. sink\nB. razor\nC. shaving foam\nD. faucet", "choices": ["A. sink", "B. razor", "C. shaving foam", "D. faucet"], "answer": "A. sink", "target": "bottle", "split": "validation"}
{"id": "7bc286c1b6_2", "scene_id": "7bc286c1b6", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 next to the faucet, about 7 cm away. It is further to the side of and below the shaving foam, about 15 cm away.", "target": "bottle", "split": "validation"}
{"id": "7bc286c1b6_3", "scene_id": "7bc286c1b6", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 3.6}, {\"object\": \"faucet\", \"relation\": \"next_to\", \"distance_cm\": 6.6}]}", "target": "bottle", "split": "validation"}
{"id": "7bc286c1b6_4", "scene_id": "7bc286c1b6", "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": "Yes", "target": "razor", "split": "validation"}
{"id": "7bc286c1b6_5", "scene_id": "7bc286c1b6", "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 razor in this scene?\nA. cloth\nB. faucet\nC. shower wall\nD. hand shower", "choices": ["A. cloth", "B. faucet", "C. shower wall", "D. hand shower"], "answer": "B. faucet", "target": "razor", "split": "validation"}
{"id": "7bc286c1b6_6", "scene_id": "7bc286c1b6", "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 razor in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The razor is to the side of and below the faucet, about 6 cm away. It is also next to the cloth, about 13 cm away. It is further to the side of and below the shower wall, about 13 cm away.", "target": "razor", "split": "validation"}
{"id": "7bc286c1b6_7", "scene_id": "7bc286c1b6", "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 razor in structured JSON format.", "choices": null, "answer": "{\"target\": \"razor\", \"support_surface\": \"bathroom counter\", \"references\": [{\"object\": \"faucet\", \"relation\": \"side_below\", \"distance_cm\": 6.1}, {\"object\": \"cloth\", \"relation\": \"next_to\", \"distance_cm\": 13.0}]}", "target": "razor", "split": "validation"}
{"id": "7bc286c1b6_8", "scene_id": "7bc286c1b6", "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 game controller appear?", "choices": null, "answer": "No", "target": "game controller", "split": "validation"}
{"id": "7bc286c1b6_9", "scene_id": "7bc286c1b6", "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. towel\nB. toilet flush button\nC. bathroom counter\nD. recessed shelve", "choices": ["A. towel", "B. toilet flush button", "C. bathroom counter", "D. recessed shelve"], "answer": "D. recessed shelve", "target": "toothbrush", "split": "validation"}
{"id": "7bc286c1b6_10", "scene_id": "7bc286c1b6", "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 below the recessed shelve, about 11 cm away. It is also to the side of and below the towel, about 39 cm away. It is further to the side of and above the bathroom counter, about 41 cm away.", "target": "toothbrush", "split": "validation"}
{"id": "7bc286c1b6_11", "scene_id": "7bc286c1b6", "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\": \"cabinet\", \"references\": [{\"object\": \"recessed shelve\", \"relation\": \"below\", \"distance_cm\": 10.8}, {\"object\": \"towel\", \"relation\": \"side_below\", \"distance_cm\": 39.3}]}", "target": "toothbrush", "split": "validation"}
{"id": "7c31a42404_0", "scene_id": "7c31a42404", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "7c31a42404_1", "scene_id": "7c31a42404", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. table\nB. couch\nC. chair\nD. ceiling lamp", "choices": ["A. table", "B. couch", "C. chair", "D. ceiling lamp"], "answer": "A. table", "target": "tv remote", "split": "validation"}
{"id": "7c31a42404_2", "scene_id": "7c31a42404", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 table, about 7 cm away. It is also to the side of and above the chair, about 41 cm away. It is further to the side of and below the ceiling lamp, about 50 cm away.", "target": "tv remote", "split": "validation"}
{"id": "7c31a42404_3", "scene_id": "7c31a42404", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\", \"relation\": \"side_above\", \"distance_cm\": 7.1}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 41.4}]}", "target": "tv remote", "split": "validation"}
{"id": "7f77abce34_0", "scene_id": "7f77abce34", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "7f77abce34_1", "scene_id": "7f77abce34", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. oven\nC. jar\nD. stove", "choices": ["A. kitchen cabinet", "B. oven", "C. jar", "D. stove"], "answer": "C. jar", "target": "bottle", "split": "validation"}
{"id": "7f77abce34_2", "scene_id": "7f77abce34", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 jar, about 16 cm away. It is also to the side of and above the stove, about 32 cm away. It is further to the side of and above the oven, about 49 cm away.", "target": "bottle", "split": "validation"}
{"id": "7f77abce34_3", "scene_id": "7f77abce34", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"jar\", \"relation\": \"near\", \"distance_cm\": 15.8}, {\"object\": \"stove\", \"relation\": \"side_above\", \"distance_cm\": 32.1}]}", "target": "bottle", "split": "validation"}
{"id": "7f77abce34_4", "scene_id": "7f77abce34", "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": "validation"}
{"id": "7f77abce34_5", "scene_id": "7f77abce34", "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 laptop, about 14 cm away. It is further to the side of and below the wardrobe, about 14 cm away.", "target": "mouse", "split": "validation"}
{"id": "7f77abce34_6", "scene_id": "7f77abce34", "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\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 13.8}, {\"object\": \"wardrobe\", \"relation\": \"side_below\", \"distance_cm\": 14.2}]}", "target": "mouse", "split": "validation"}
{"id": "7f77abce34_7", "scene_id": "7f77abce34", "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": "validation"}
{"id": "7f77abce34_8", "scene_id": "7f77abce34", "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. monitor\nB. laptop\nC. table\nD. office chair", "choices": ["A. monitor", "B. laptop", "C. table", "D. office chair"], "answer": "C. table", "target": "notebook", "split": "validation"}
{"id": "7f77abce34_9", "scene_id": "7f77abce34", "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 to the side of and above the table, about 5 cm away. It is also to the side of and above the laptop, about 13 cm away. It is further to the side of and above the office chair, about 34 cm away.", "target": "notebook", "split": "validation"}
{"id": "7f77abce34_10", "scene_id": "7f77abce34", "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\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 4.9}, {\"object\": \"laptop\", \"relation\": \"side_above\", \"distance_cm\": 13.0}]}", "target": "notebook", "split": "validation"}
{"id": "7f77abce34_11", "scene_id": "7f77abce34", "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 pencil case appear?", "choices": null, "answer": "No", "target": "pencil case", "split": "validation"}
{"id": "7f77abce34_12", "scene_id": "7f77abce34", "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. wardrobe\nB. mouse\nC. split\nD. monitor", "choices": ["A. wardrobe", "B. mouse", "C. split", "D. monitor"], "answer": "D. monitor", "target": "power strip", "split": "validation"}
{"id": "7f77abce34_13", "scene_id": "7f77abce34", "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 table. It is also to the side of and below the monitor, about 12 cm away. It is further near the split, about 30 cm away.", "target": "power strip", "split": "validation"}
{"id": "7f77abce34_14", "scene_id": "7f77abce34", "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\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 11.9}, {\"object\": \"split\", \"relation\": \"near\", \"distance_cm\": 29.9}]}", "target": "power strip", "split": "validation"}
{"id": "7f77abce34_15", "scene_id": "7f77abce34", "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 tote bag appear?", "choices": null, "answer": "Yes", "target": "tote bag", "split": "validation"}
{"id": "7f77abce34_16", "scene_id": "7f77abce34", "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 tote bag in this scene?\nA. office chair\nB. backpack\nC. light switch\nD. doorframe", "choices": ["A. office chair", "B. backpack", "C. light switch", "D. doorframe"], "answer": "A. office chair", "target": "tote bag", "split": "validation"}
{"id": "7f77abce34_17", "scene_id": "7f77abce34", "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 tote bag in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tote bag is on the desk. It is also to the side of and above the office chair, about 14 cm away. It is further near the backpack, about 17 cm away.", "target": "tote bag", "split": "validation"}
{"id": "7f77abce34_18", "scene_id": "7f77abce34", "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 tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 14.4}, {\"object\": \"backpack\", \"relation\": \"near\", \"distance_cm\": 16.9}]}", "target": "tote bag", "split": "validation"}
{"id": "8890d0a267_0", "scene_id": "8890d0a267", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "8890d0a267_1", "scene_id": "8890d0a267", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. bed\nB. blanket\nC. bottle\nD. bag", "choices": ["A. bed", "B. blanket", "C. bottle", "D. bag"], "answer": "C. bottle", "target": "backpack", "split": "validation"}
{"id": "8890d0a267_2", "scene_id": "8890d0a267", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 bottle, about 28 cm away.", "target": "backpack", "split": "validation"}
{"id": "8890d0a267_3", "scene_id": "8890d0a267", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 27.6}]}", "target": "backpack", "split": "validation"}
{"id": "891865308a_0", "scene_id": "891865308a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "891865308a_1", "scene_id": "891865308a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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 socket\nB. bed\nC. folded blanket\nD. blanket", "choices": ["A. power socket", "B. bed", "C. folded blanket", "D. blanket"], "answer": "C. folded blanket", "target": "extension cord", "split": "validation"}
{"id": "891865308a_2", "scene_id": "891865308a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 folded blanket, about 50 cm away. It is also to the side of and below the power socket, about 72 cm away.", "target": "extension cord", "split": "validation"}
{"id": "891865308a_3", "scene_id": "891865308a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"bed\", \"references\": [{\"object\": \"folded blanket\", \"relation\": \"side_below\", \"distance_cm\": 50.4}, {\"object\": \"power socket\", \"relation\": \"side_below\", \"distance_cm\": 72.4}]}", "target": "extension cord", "split": "validation"}
{"id": "891865308a_4", "scene_id": "891865308a", "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": "validation"}
{"id": "891865308a_5", "scene_id": "891865308a", "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. mouse\nB. lotion bottle\nC. chair\nD. monitor", "choices": ["A. mouse", "B. lotion bottle", "C. chair", "D. monitor"], "answer": "D. monitor", "target": "headphones", "split": "validation"}
{"id": "891865308a_6", "scene_id": "891865308a", "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 below the monitor, about 8 cm away. It is further near the mouse, about 15 cm away.", "target": "headphones", "split": "validation"}
{"id": "891865308a_7", "scene_id": "891865308a", "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\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 8.5}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 15.2}]}", "target": "headphones", "split": "validation"}
{"id": "891865308a_8", "scene_id": "891865308a", "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 purse appear?", "choices": null, "answer": "No", "target": "purse", "split": "validation"}
{"id": "891865308a_9", "scene_id": "891865308a", "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 lotion bottle in this scene?\nA. mouse\nB. headphones\nC. monitor\nD. chair", "choices": ["A. mouse", "B. headphones", "C. monitor", "D. chair"], "answer": "B. headphones", "target": "lotion bottle", "split": "validation"}
{"id": "891865308a_10", "scene_id": "891865308a", "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 lotion bottle in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The lotion bottle is on the desk. It is also near the headphones, about 25 cm away. It is further near the mouse, about 35 cm away.", "target": "lotion bottle", "split": "validation"}
{"id": "891865308a_11", "scene_id": "891865308a", "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 lotion bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"lotion bottle\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"headphones\", \"relation\": \"near\", \"distance_cm\": 25.1}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 35.1}]}", "target": "lotion bottle", "split": "validation"}
{"id": "891865308a_12", "scene_id": "891865308a", "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": "validation"}
{"id": "891865308a_13", "scene_id": "891865308a", "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. headphones\nC. chair\nD. lotion bottle", "choices": ["A. monitor", "B. headphones", "C. chair", "D. lotion bottle"], "answer": "B. headphones", "target": "mouse", "split": "validation"}
{"id": "891865308a_14", "scene_id": "891865308a", "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 near the headphones, about 15 cm away. It is further to the side of and above the chair, about 23 cm away.", "target": "mouse", "split": "validation"}
{"id": "891865308a_15", "scene_id": "891865308a", "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\": \"headphones\", \"relation\": \"near\", \"distance_cm\": 15.2}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 23.4}]}", "target": "mouse", "split": "validation"}
{"id": "893fb90e89_0", "scene_id": "893fb90e89", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "893fb90e89_1", "scene_id": "893fb90e89", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. binder\nB. bag\nC. books\nD. table", "choices": ["A. binder", "B. bag", "C. books", "D. table"], "answer": "D. table", "target": "backpack", "split": "validation"}
{"id": "893fb90e89_2", "scene_id": "893fb90e89", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 18 cm away.", "target": "backpack", "split": "validation"}
{"id": "893fb90e89_3", "scene_id": "893fb90e89", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 17.7}]}", "target": "backpack", "split": "validation"}
{"id": "893fb90e89_4", "scene_id": "893fb90e89", "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": "validation"}
{"id": "893fb90e89_5", "scene_id": "893fb90e89", "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. pencil\nB. split\nC. office chair\nD. tray", "choices": ["A. pencil", "B. split", "C. office chair", "D. tray"], "answer": "A. pencil", "target": "notebook", "split": "validation"}
{"id": "893fb90e89_6", "scene_id": "893fb90e89", "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 near the pencil, about 16 cm away. It is further to the side of and above the office chair, about 37 cm away.", "target": "notebook", "split": "validation"}
{"id": "893fb90e89_7", "scene_id": "893fb90e89", "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\": \"pencil\", \"relation\": \"near\", \"distance_cm\": 16.2}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 36.6}]}", "target": "notebook", "split": "validation"}
{"id": "893fb90e89_8", "scene_id": "893fb90e89", "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": "validation"}
{"id": "893fb90e89_9", "scene_id": "893fb90e89", "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. poster\nB. table\nC. box\nD. file holder", "choices": ["A. poster", "B. table", "C. box", "D. file holder"], "answer": "C. box", "target": "pen holder", "split": "validation"}
{"id": "893fb90e89_10", "scene_id": "893fb90e89", "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 cabinet. It is also next to the box, about 6 cm away. It is further to the side of and above the table, about 11 cm away.", "target": "pen holder", "split": "validation"}
{"id": "893fb90e89_11", "scene_id": "893fb90e89", "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\": \"cabinet\", \"references\": [{\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 5.9}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 11.1}]}", "target": "pen holder", "split": "validation"}
{"id": "893fb90e89_12", "scene_id": "893fb90e89", "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 appear?", "choices": null, "answer": "Yes", "target": "pencil", "split": "validation"}
{"id": "893fb90e89_13", "scene_id": "893fb90e89", "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 in this scene?\nA. notebook\nB. office chair\nC. split\nD. tray", "choices": ["A. notebook", "B. office chair", "C. split", "D. tray"], "answer": "A. notebook", "target": "pencil", "split": "validation"}
{"id": "893fb90e89_14", "scene_id": "893fb90e89", "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 in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The pencil is on the table. It is also near the notebook, about 16 cm away. It is further near the split, about 28 cm away.", "target": "pencil", "split": "validation"}
{"id": "893fb90e89_15", "scene_id": "893fb90e89", "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 in structured JSON format.", "choices": null, "answer": "{\"target\": \"pencil\", \"support_surface\": \"table\", \"references\": [{\"object\": \"notebook\", \"relation\": \"near\", \"distance_cm\": 16.2}, {\"object\": \"split\", \"relation\": \"near\", \"distance_cm\": 28.3}]}", "target": "pencil", "split": "validation"}
{"id": "8b5caf3398_0", "scene_id": "8b5caf3398", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "8b5caf3398_1", "scene_id": "8b5caf3398", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. window frame\nC. cabinet\nD. tissue box", "choices": ["A. paper", "B. window frame", "C. cabinet", "D. tissue box"], "answer": "B. window frame", "target": "bottle", "split": "validation"}
{"id": "8b5caf3398_2", "scene_id": "8b5caf3398", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 frame, about 8 cm away. It is also to the side of and below the window, about 32 cm away. It is further to the side of and above the tissue box, about 83 cm away.", "target": "bottle", "split": "validation"}
{"id": "8b5caf3398_3", "scene_id": "8b5caf3398", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"window frame\", \"relation\": \"side_below\", \"distance_cm\": 8.0}, {\"object\": \"window\", \"relation\": \"side_below\", \"distance_cm\": 31.5}]}", "target": "bottle", "split": "validation"}
{"id": "8b5caf3398_4", "scene_id": "8b5caf3398", "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": "validation"}
{"id": "8b5caf3398_5", "scene_id": "8b5caf3398", "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. cabinet\nB. bottle\nC. window frame\nD. paper", "choices": ["A. cabinet", "B. bottle", "C. window frame", "D. paper"], "answer": "D. paper", "target": "tissue box", "split": "validation"}
{"id": "8b5caf3398_6", "scene_id": "8b5caf3398", "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 cabinet. It is also to the side of and above the paper, about 5 cm away. It is further to the side of and below the bottle, about 83 cm away.", "target": "tissue box", "split": "validation"}
{"id": "8b5caf3398_7", "scene_id": "8b5caf3398", "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\": \"cabinet\", \"references\": [{\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 4.7}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 82.8}]}", "target": "tissue box", "split": "validation"}
{"id": "8be0cd3817_0", "scene_id": "8be0cd3817", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "8be0cd3817_1", "scene_id": "8be0cd3817", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. pan\nB. kitchen counter\nC. dish rack\nD. power socket", "choices": ["A. pan", "B. kitchen counter", "C. dish rack", "D. power socket"], "answer": "D. power socket", "target": "bottle", "split": "validation"}
{"id": "8be0cd3817_2", "scene_id": "8be0cd3817", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 power socket, about 28 cm away. It is further to the side of and above the pan, about 37 cm away.", "target": "bottle", "split": "validation"}
{"id": "8be0cd3817_3", "scene_id": "8be0cd3817", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"power socket\", \"relation\": \"side_above\", \"distance_cm\": 27.8}, {\"object\": \"pan\", \"relation\": \"side_above\", \"distance_cm\": 37.1}]}", "target": "bottle", "split": "validation"}
{"id": "8be0cd3817_4", "scene_id": "8be0cd3817", "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": "validation"}
{"id": "8be0cd3817_5", "scene_id": "8be0cd3817", "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 earbuds in this scene?\nA. blanket\nB. laptop\nC. shelf\nD. table lamp", "choices": ["A. blanket", "B. laptop", "C. shelf", "D. table lamp"], "answer": "C. shelf", "target": "earbuds", "split": "validation"}
{"id": "8be0cd3817_6", "scene_id": "8be0cd3817", "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 earbuds in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The earbuds is above the shelf, about 3 cm away. It is also next to the laptop, about 5 cm away. It is further to the side of and below the table lamp, about 8 cm away.", "target": "earbuds", "split": "validation"}
{"id": "8be0cd3817_7", "scene_id": "8be0cd3817", "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 earbuds in structured JSON format.", "choices": null, "answer": "{\"target\": \"earbuds\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"shelf\", \"relation\": \"above\", \"distance_cm\": 2.8}, {\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 5.4}]}", "target": "earbuds", "split": "validation"}
{"id": "8be0cd3817_8", "scene_id": "8be0cd3817", "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": "validation"}
{"id": "8be0cd3817_9", "scene_id": "8be0cd3817", "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. power strip\nB. wardrobe\nC. office chair\nD. monitor", "choices": ["A. power strip", "B. wardrobe", "C. office chair", "D. monitor"], "answer": "D. monitor", "target": "keyboard", "split": "validation"}
{"id": "8be0cd3817_10", "scene_id": "8be0cd3817", "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 6 cm away. It is further to the side of and above the office chair, about 21 cm away.", "target": "keyboard", "split": "validation"}
{"id": "8be0cd3817_11", "scene_id": "8be0cd3817", "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\": 5.7}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 21.2}]}", "target": "keyboard", "split": "validation"}
{"id": "8be0cd3817_12", "scene_id": "8be0cd3817", "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 cord appear?", "choices": null, "answer": "No", "target": "power cord", "split": "validation"}
{"id": "8be0cd3817_13", "scene_id": "8be0cd3817", "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. monitor\nC. kettle\nD. toaster", "choices": ["A. keyboard", "B. monitor", "C. kettle", "D. toaster"], "answer": "B. monitor", "target": "power strip", "split": "validation"}
{"id": "8be0cd3817_14", "scene_id": "8be0cd3817", "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 to the side of and below the monitor, about 14 cm away. It is further to the side of and below the kettle, about 26 cm away.", "target": "power strip", "split": "validation"}
{"id": "8be0cd3817_15", "scene_id": "8be0cd3817", "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\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 13.8}, {\"object\": \"kettle\", \"relation\": \"side_below\", \"distance_cm\": 26.2}]}", "target": "power strip", "split": "validation"}
{"id": "8be0cd3817_16", "scene_id": "8be0cd3817", "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 deodorant bottle appear?", "choices": null, "answer": "No", "target": "deodorant bottle", "split": "validation"}
{"id": "8be0cd3817_17", "scene_id": "8be0cd3817", "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 remote control in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The remote control is next to the powerbank, about 8 cm away. It is also to the side of and below the blanket, about 8 cm away. It is further near the dvd player, about 18 cm away.", "target": "remote control", "split": "validation"}
{"id": "8be0cd3817_18", "scene_id": "8be0cd3817", "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 remote control in structured JSON format.", "choices": null, "answer": "{\"target\": \"remote control\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"powerbank\", \"relation\": \"next_to\", \"distance_cm\": 8.3}, {\"object\": \"blanket\", \"relation\": \"side_below\", \"distance_cm\": 8.3}]}", "target": "remote control", "split": "validation"}
{"id": "8fc40ba77b_0", "scene_id": "8fc40ba77b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "8fc40ba77b_1", "scene_id": "8fc40ba77b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. heater\nC. chair\nD. power socket", "choices": ["A. windowsill", "B. heater", "C. chair", "D. power socket"], "answer": "B. heater", "target": "power strip", "split": "validation"}
{"id": "8fc40ba77b_2", "scene_id": "8fc40ba77b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 heater, about 11 cm away. It is also to the side of and below the power socket, about 19 cm away. It is further to the side of and below the windowsill, about 64 cm away.", "target": "power strip", "split": "validation"}
{"id": "8fc40ba77b_3", "scene_id": "8fc40ba77b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"heater\", \"relation\": \"side_below\", \"distance_cm\": 10.7}, {\"object\": \"power socket\", \"relation\": \"side_below\", \"distance_cm\": 18.6}]}", "target": "power strip", "split": "validation"}
{"id": "927aacd5d1_0", "scene_id": "927aacd5d1", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "927aacd5d1_1", "scene_id": "927aacd5d1", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. box\nC. pillow\nD. table lamp", "choices": ["A. power socket", "B. box", "C. pillow", "D. table lamp"], "answer": "D. table lamp", "target": "cup", "split": "validation"}
{"id": "927aacd5d1_2", "scene_id": "927aacd5d1", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 nightstand. It is also to the side of and below the table lamp, about 7 cm away. It is further next to the power socket, about 10 cm away.", "target": "cup", "split": "validation"}
{"id": "927aacd5d1_3", "scene_id": "927aacd5d1", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"nightstand\", \"references\": [{\"object\": \"table lamp\", \"relation\": \"side_below\", \"distance_cm\": 6.5}, {\"object\": \"power socket\", \"relation\": \"next_to\", \"distance_cm\": 10.1}]}", "target": "cup", "split": "validation"}
{"id": "927aacd5d1_4", "scene_id": "927aacd5d1", "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": "validation"}
{"id": "927aacd5d1_5", "scene_id": "927aacd5d1", "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. webcam\nB. cap\nC. box\nD. clothes", "choices": ["A. webcam", "B. cap", "C. box", "D. clothes"], "answer": "B. cap", "target": "spray bottle", "split": "validation"}
{"id": "927aacd5d1_6", "scene_id": "927aacd5d1", "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 next to the cap. It is further near the webcam, about 31 cm away.", "target": "spray bottle", "split": "validation"}
{"id": "927aacd5d1_7", "scene_id": "927aacd5d1", "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\": \"cap\", \"relation\": \"next_to\", \"distance_cm\": 2.4}, {\"object\": \"webcam\", \"relation\": \"near\", \"distance_cm\": 31.0}]}", "target": "spray bottle", "split": "validation"}
{"id": "96b7c379eb_0", "scene_id": "96b7c379eb", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "96b7c379eb_1", "scene_id": "96b7c379eb", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. toilet flush\nB. toilet\nC. power sockets\nD. shelf", "choices": ["A. toilet flush", "B. toilet", "C. power sockets", "D. shelf"], "answer": "A. toilet flush", "target": "spray bottle", "split": "validation"}
{"id": "96b7c379eb_2", "scene_id": "96b7c379eb", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 above the toilet flush, about 9 cm away. It is also to the side of and above the power sockets, about 25 cm away. It is further near the shelf, about 26 cm away.", "target": "spray bottle", "split": "validation"}
{"id": "96b7c379eb_3", "scene_id": "96b7c379eb", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"toilet flush\", \"relation\": \"above\", \"distance_cm\": 8.9}, {\"object\": \"power sockets\", \"relation\": \"side_above\", \"distance_cm\": 24.8}]}", "target": "spray bottle", "split": "validation"}
{"id": "9816c49e97_0", "scene_id": "9816c49e97", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "9816c49e97_1", "scene_id": "9816c49e97", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. chair\nB. table\nC. monitor\nD. bag", "choices": ["A. chair", "B. table", "C. monitor", "D. bag"], "answer": "B. table", "target": "water bottle", "split": "validation"}
{"id": "9816c49e97_2", "scene_id": "9816c49e97", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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, about 8 cm away. It is also to the side of and above the chair, about 16 cm away.", "target": "water bottle", "split": "validation"}
{"id": "9816c49e97_3", "scene_id": "9816c49e97", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 8.3}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 16.2}]}", "target": "water bottle", "split": "validation"}
{"id": "99a2503d63_0", "scene_id": "99a2503d63", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "99a2503d63_1", "scene_id": "99a2503d63", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. microwave\nC. table\nD. bowl", "choices": ["A. chair", "B. microwave", "C. table", "D. bowl"], "answer": "A. chair", "target": "power strip", "split": "validation"}
{"id": "99a2503d63_2", "scene_id": "99a2503d63", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 5 cm away. It is also near the table, about 22 cm away. It is further to the side of and below the microwave, about 28 cm away.", "target": "power strip", "split": "validation"}
{"id": "99a2503d63_3", "scene_id": "99a2503d63", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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_below\", \"distance_cm\": 5.1}, {\"object\": \"table\", \"relation\": \"near\", \"distance_cm\": 22.0}]}", "target": "power strip", "split": "validation"}
{"id": "99fa5c25e1_0", "scene_id": "99fa5c25e1", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "99fa5c25e1_1", "scene_id": "99fa5c25e1", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. hammer holder\nB. papers\nC. cloth\nD. paper holder", "choices": ["A. hammer holder", "B. papers", "C. cloth", "D. paper holder"], "answer": "B. papers", "target": "notebook", "split": "validation"}
{"id": "99fa5c25e1_2", "scene_id": "99fa5c25e1", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 papers, about 12 cm away. It is also near the cloth, about 16 cm away. It is further near the hammer holder, about 20 cm away.", "target": "notebook", "split": "validation"}
{"id": "99fa5c25e1_3", "scene_id": "99fa5c25e1", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"papers\", \"references\": [{\"object\": \"papers\", \"relation\": \"next_to\", \"distance_cm\": 12.2}, {\"object\": \"cloth\", \"relation\": \"near\", \"distance_cm\": 16.0}]}", "target": "notebook", "split": "validation"}
{"id": "9e9d017ad1_0", "scene_id": "9e9d017ad1", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "9e9d017ad1_1", "scene_id": "9e9d017ad1", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. sink\nB. soap dispenser\nC. cabinet\nD. monitor", "choices": ["A. sink", "B. soap dispenser", "C. cabinet", "D. monitor"], "answer": "A. sink", "target": "bottle", "split": "validation"}
{"id": "9e9d017ad1_2", "scene_id": "9e9d017ad1", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 sink, about 5 cm away. It is further to the side of and below the soap dispenser, about 63 cm away.", "target": "bottle", "split": "validation"}
{"id": "9e9d017ad1_3", "scene_id": "9e9d017ad1", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"sink\", \"relation\": \"next_to\", \"distance_cm\": 4.9}, {\"object\": \"soap dispenser\", \"relation\": \"side_below\", \"distance_cm\": 63.1}]}", "target": "bottle", "split": "validation"}
{"id": "9f139a318d_0", "scene_id": "9f139a318d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "9f139a318d_1", "scene_id": "9f139a318d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. keyboard\nC. telephone\nD. paper", "choices": ["A. book", "B. keyboard", "C. telephone", "D. paper"], "answer": "B. keyboard", "target": "mouse", "split": "validation"}
{"id": "9f139a318d_2", "scene_id": "9f139a318d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 telephone, about 7 cm away.", "target": "mouse", "split": "validation"}
{"id": "9f139a318d_3", "scene_id": "9f139a318d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"telephone\", \"relation\": \"next_to\", \"distance_cm\": 7.1}]}", "target": "mouse", "split": "validation"}
{"id": "9f139a318d_4", "scene_id": "9f139a318d", "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": "validation"}
{"id": "9f139a318d_5", "scene_id": "9f139a318d", "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. cabinet\nB. backpack\nC. table\nD. jacket", "choices": ["A. cabinet", "B. backpack", "C. table", "D. jacket"], "answer": "A. cabinet", "target": "power strip", "split": "validation"}
{"id": "9f139a318d_6", "scene_id": "9f139a318d", "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 cabinet, about 7 cm away. It is also to the side of and below the jacket, about 46 cm away. It is further to the side of and below the backpack, about 56 cm away.", "target": "power strip", "split": "validation"}
{"id": "9f139a318d_7", "scene_id": "9f139a318d", "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\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 7.0}, {\"object\": \"jacket\", \"relation\": \"side_below\", \"distance_cm\": 46.5}]}", "target": "power strip", "split": "validation"}
{"id": "a1d9da703c_0", "scene_id": "a1d9da703c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "a1d9da703c_1", "scene_id": "a1d9da703c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. pen\nB. monitor\nC. box\nD. chair", "choices": ["A. pen", "B. monitor", "C. box", "D. chair"], "answer": "B. monitor", "target": "mug", "split": "validation"}
{"id": "a1d9da703c_2", "scene_id": "a1d9da703c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 monitor, about 4 cm away. It is further next to the box, about 6 cm away.", "target": "mug", "split": "validation"}
{"id": "a1d9da703c_3", "scene_id": "a1d9da703c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"under\", \"distance_cm\": 3.9}, {\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 6.2}]}", "target": "mug", "split": "validation"}
{"id": "a1d9da703c_4", "scene_id": "a1d9da703c", "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": "validation"}
{"id": "a1d9da703c_5", "scene_id": "a1d9da703c", "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. mouse\nB. chair\nC. mug\nD. box", "choices": ["A. mouse", "B. chair", "C. mug", "D. box"], "answer": "B. chair", "target": "pen", "split": "validation"}
{"id": "a1d9da703c_6", "scene_id": "a1d9da703c", "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 chair, about 10 cm away. It is further next to the mouse, about 12 cm away.", "target": "pen", "split": "validation"}
{"id": "a1d9da703c_7", "scene_id": "a1d9da703c", "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\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 9.8}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 12.4}]}", "target": "pen", "split": "validation"}
{"id": "a1d9da703c_8", "scene_id": "a1d9da703c", "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": "validation"}
{"id": "a1d9da703c_9", "scene_id": "a1d9da703c", "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. box\nB. mug\nC. keyboard\nD. monitor", "choices": ["A. box", "B. mug", "C. keyboard", "D. monitor"], "answer": "A. box", "target": "stapler", "split": "validation"}
{"id": "a1d9da703c_10", "scene_id": "a1d9da703c", "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 table. It is also near the box, about 35 cm away. It is further near the keyboard, about 38 cm away.", "target": "stapler", "split": "validation"}
{"id": "a1d9da703c_11", "scene_id": "a1d9da703c", "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\": \"box\", \"relation\": \"near\", \"distance_cm\": 35.2}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 38.0}]}", "target": "stapler", "split": "validation"}
{"id": "a30b666f83_0", "scene_id": "a30b666f83", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "a30b666f83_1", "scene_id": "a30b666f83", "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 basket, about 3 cm away. It is also next to the dish soap bottle, about 3 cm away. It is further to the side of and below the power outlet, about 22 cm away.", "target": "spray bottle", "split": "validation"}
{"id": "a30b666f83_2", "scene_id": "a30b666f83", "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\": \"basket\", \"references\": [{\"object\": \"basket\", \"relation\": \"side_above\", \"distance_cm\": 2.6}, {\"object\": \"dish soap bottle\", \"relation\": \"next_to\", \"distance_cm\": 3.4}]}", "target": "spray bottle", "split": "validation"}
{"id": "a492fe77aa_0", "scene_id": "a492fe77aa", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "a492fe77aa_1", "scene_id": "a492fe77aa", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. compact wardrobe\nB. paper bag\nC. clothes\nD. picture frame", "choices": ["A. compact wardrobe", "B. paper bag", "C. clothes", "D. picture frame"], "answer": "C. clothes", "target": "backpack", "split": "validation"}
{"id": "a492fe77aa_2", "scene_id": "a492fe77aa", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 picture frame, about 11 cm away. It is also to the side of and below the paper bag, about 39 cm away.", "target": "backpack", "split": "validation"}
{"id": "a492fe77aa_3", "scene_id": "a492fe77aa", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"paper bag\", \"references\": [{\"object\": \"picture frame\", \"relation\": \"side_below\", \"distance_cm\": 10.8}, {\"object\": \"paper bag\", \"relation\": \"side_below\", \"distance_cm\": 39.4}]}", "target": "backpack", "split": "validation"}
{"id": "a492fe77aa_4", "scene_id": "a492fe77aa", "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": "validation"}
{"id": "a492fe77aa_5", "scene_id": "a492fe77aa", "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. clothes\nB. backpack\nC. table\nD. plant pot", "choices": ["A. clothes", "B. backpack", "C. table", "D. plant pot"], "answer": "C. table", "target": "extension cord", "split": "validation"}
{"id": "a492fe77aa_6", "scene_id": "a492fe77aa", "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 table, about 23 cm away. It is also to the side of and below the plant pot, about 38 cm away.", "target": "extension cord", "split": "validation"}
{"id": "a492fe77aa_7", "scene_id": "a492fe77aa", "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\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 23.4}, {\"object\": \"plant pot\", \"relation\": \"side_below\", \"distance_cm\": 38.4}]}", "target": "extension cord", "split": "validation"}
{"id": "a4e227f506_0", "scene_id": "a4e227f506", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "a4e227f506_1", "scene_id": "a4e227f506", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. cable raceway\nB. eraser\nC. computer tower\nD. power socket", "choices": ["A. cable raceway", "B. eraser", "C. computer tower", "D. power socket"], "answer": "A. cable raceway", "target": "bottle", "split": "validation"}
{"id": "a4e227f506_2", "scene_id": "a4e227f506", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 power socket, about 18 cm away. It is also to the side of and above the computer tower, about 69 cm away. It is further to the side of and above the eraser, about 78 cm away.", "target": "bottle", "split": "validation"}
{"id": "a4e227f506_3", "scene_id": "a4e227f506", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"cable raceway\", \"references\": [{\"object\": \"power socket\", \"relation\": \"side_above\", \"distance_cm\": 17.9}, {\"object\": \"computer tower\", \"relation\": \"side_above\", \"distance_cm\": 69.0}]}", "target": "bottle", "split": "validation"}
{"id": "a4e227f506_4", "scene_id": "a4e227f506", "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": "validation"}
{"id": "a4e227f506_5", "scene_id": "a4e227f506", "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 eraser appear?", "choices": null, "answer": "Yes", "target": "eraser", "split": "validation"}
{"id": "a4e227f506_6", "scene_id": "a4e227f506", "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 eraser in this scene?\nA. cable raceway\nB. refrigerator\nC. power socket\nD. monitor", "choices": ["A. cable raceway", "B. refrigerator", "C. power socket", "D. monitor"], "answer": "A. cable raceway", "target": "eraser", "split": "validation"}
{"id": "a4e227f506_7", "scene_id": "a4e227f506", "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 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 cable raceway, about 15 cm away. It is also to the side of and above the power socket, about 18 cm away. It is further to the side of and above the refrigerator, about 42 cm away.", "target": "eraser", "split": "validation"}
{"id": "a4e227f506_8", "scene_id": "a4e227f506", "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 eraser in structured JSON format.", "choices": null, "answer": "{\"target\": \"eraser\", \"support_surface\": \"power socket\", \"references\": [{\"object\": \"cable raceway\", \"relation\": \"side_above\", \"distance_cm\": 15.2}, {\"object\": \"power socket\", \"relation\": \"side_above\", \"distance_cm\": 18.4}]}", "target": "eraser", "split": "validation"}
{"id": "a4e227f506_9", "scene_id": "a4e227f506", "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 earphones appear?", "choices": null, "answer": "No", "target": "earphones", "split": "validation"}
{"id": "a4e227f506_10", "scene_id": "a4e227f506", "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 keyboard in this scene?\nA. computer tower\nB. mouse\nC. monitor\nD. printer", "choices": ["A. computer tower", "B. mouse", "C. monitor", "D. printer"], "answer": "C. monitor", "target": "keyboard", "split": "validation"}
{"id": "a4e227f506_11", "scene_id": "a4e227f506", "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 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 next to the mouse, about 6 cm away. It is further to the side of and below the computer tower, about 23 cm away.", "target": "keyboard", "split": "validation"}
{"id": "a4e227f506_12", "scene_id": "a4e227f506", "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 keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 5.0}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 6.5}]}", "target": "keyboard", "split": "validation"}
{"id": "a4e227f506_13", "scene_id": "a4e227f506", "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 tape dispenser appear?", "choices": null, "answer": "Yes", "target": "tape dispenser", "split": "validation"}
{"id": "a4e227f506_14", "scene_id": "a4e227f506", "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 tape dispenser in this scene?\nA. chair\nB. monitor\nC. computer tower\nD. partition", "choices": ["A. chair", "B. monitor", "C. computer tower", "D. partition"], "answer": "A. chair", "target": "tape dispenser", "split": "validation"}
{"id": "a4e227f506_15", "scene_id": "a4e227f506", "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 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 chair, about 18 cm away. It is also to the side of and above the computer tower, about 33 cm away. It is further to the side of and below the monitor, about 46 cm away.", "target": "tape dispenser", "split": "validation"}
{"id": "a4e227f506_16", "scene_id": "a4e227f506", "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 tape dispenser in structured JSON format.", "choices": null, "answer": "{\"target\": \"tape dispenser\", \"support_surface\": \"table\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 18.1}, {\"object\": \"computer tower\", \"relation\": \"side_above\", \"distance_cm\": 32.7}]}", "target": "tape dispenser", "split": "validation"}
{"id": "a4f10da05d_0", "scene_id": "a4f10da05d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "a4f10da05d_1", "scene_id": "a4f10da05d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. light switches\nC. reusable shopping bag\nD. doorframe", "choices": ["A. jacket", "B. light switches", "C. reusable shopping bag", "D. doorframe"], "answer": "C. reusable shopping bag", "target": "backpack", "split": "validation"}
{"id": "a4f10da05d_2", "scene_id": "a4f10da05d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 jacket, about 11 cm away. It is further near the doorframe, about 17 cm away.", "target": "backpack", "split": "validation"}
{"id": "a4f10da05d_3", "scene_id": "a4f10da05d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"jacket\", \"relation\": \"side_below\", \"distance_cm\": 10.8}, {\"object\": \"doorframe\", \"relation\": \"near\", \"distance_cm\": 16.8}]}", "target": "backpack", "split": "validation"}
{"id": "a4f10da05d_4", "scene_id": "a4f10da05d", "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": "validation"}
{"id": "a4f10da05d_5", "scene_id": "a4f10da05d", "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 utility cart, about 8 cm away. It is also to the side of and below the cabinet, about 9 cm away. It is further to the side of and below the cosmetic bottle, about 47 cm away.", "target": "cable", "split": "validation"}
{"id": "a4f10da05d_6", "scene_id": "a4f10da05d", "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\": \"cabinet\", \"references\": [{\"object\": \"utility cart\", \"relation\": \"side_below\", \"distance_cm\": 8.5}, {\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 8.6}]}", "target": "cable", "split": "validation"}
{"id": "a4f10da05d_7", "scene_id": "a4f10da05d", "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 bottle appear?", "choices": null, "answer": "No", "target": "bottle", "split": "validation"}
{"id": "a4f10da05d_8", "scene_id": "a4f10da05d", "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. notebook\nC. mouse\nD. reusable cup", "choices": ["A. monitor", "B. notebook", "C. mouse", "D. reusable cup"], "answer": "C. mouse", "target": "keyboard", "split": "validation"}
{"id": "a4f10da05d_9", "scene_id": "a4f10da05d", "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 next to the mouse, about 4 cm away. It is further to the side of and below the reusable cup, about 6 cm away.", "target": "keyboard", "split": "validation"}
{"id": "a4f10da05d_10", "scene_id": "a4f10da05d", "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\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 4.4}, {\"object\": \"reusable cup\", \"relation\": \"side_below\", \"distance_cm\": 6.2}]}", "target": "keyboard", "split": "validation"}
{"id": "a4f10da05d_11", "scene_id": "a4f10da05d", "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 plug appear?", "choices": null, "answer": "No", "target": "power plug", "split": "validation"}
{"id": "a4f10da05d_12", "scene_id": "a4f10da05d", "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. clothes drying rack\nC. cloth\nD. reusable cup", "choices": ["A. keyboard", "B. clothes drying rack", "C. cloth", "D. reusable cup"], "answer": "A. keyboard", "target": "mouse", "split": "validation"}
{"id": "a4f10da05d_13", "scene_id": "a4f10da05d", "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 desk. It is also next to the keyboard, about 4 cm away. It is further to the side of and below the cloth, about 10 cm away.", "target": "mouse", "split": "validation"}
{"id": "a4f10da05d_14", "scene_id": "a4f10da05d", "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\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 4.4}, {\"object\": \"cloth\", \"relation\": \"side_below\", \"distance_cm\": 9.7}]}", "target": "mouse", "split": "validation"}
{"id": "a4f10da05d_15", "scene_id": "a4f10da05d", "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 notebook appear?", "choices": null, "answer": "Yes", "target": "notebook", "split": "validation"}
{"id": "a4f10da05d_16", "scene_id": "a4f10da05d", "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 notebook in this scene?\nA. keyboard\nB. books\nC. reusable cup\nD. curtain", "choices": ["A. keyboard", "B. books", "C. reusable cup", "D. curtain"], "answer": "B. books", "target": "notebook", "split": "validation"}
{"id": "a4f10da05d_17", "scene_id": "a4f10da05d", "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 notebook in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The notebook is on the desk. It is also under the books, about 3 cm away. It is further to the side of and below the curtain, about 7 cm away.", "target": "notebook", "split": "validation"}
{"id": "a4f10da05d_18", "scene_id": "a4f10da05d", "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 notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"books\", \"relation\": \"under\", \"distance_cm\": 2.6}, {\"object\": \"curtain\", \"relation\": \"side_below\", \"distance_cm\": 7.5}]}", "target": "notebook", "split": "validation"}
{"id": "a4f10da05d_19", "scene_id": "a4f10da05d", "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 case appear?", "choices": null, "answer": "Yes", "target": "pencil case", "split": "validation"}
{"id": "a4f10da05d_20", "scene_id": "a4f10da05d", "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 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 under the reusable cup, about 4 cm away. It is further under the table lamp, about 4 cm away.", "target": "pencil case", "split": "validation"}
{"id": "a4f10da05d_21", "scene_id": "a4f10da05d", "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 pencil case in structured JSON format.", "choices": null, "answer": "{\"target\": \"pencil case\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"reusable cup\", \"relation\": \"under\", \"distance_cm\": 3.9}, {\"object\": \"table lamp\", \"relation\": \"under\", \"distance_cm\": 3.9}]}", "target": "pencil case", "split": "validation"}
{"id": "a7631adbe7_0", "scene_id": "a7631adbe7", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "a7631adbe7_1", "scene_id": "a7631adbe7", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. box\nB. chair\nC. coffee table\nD. coffee maker", "choices": ["A. box", "B. chair", "C. coffee table", "D. coffee maker"], "answer": "B. chair", "target": "mug", "split": "validation"}
{"id": "a7631adbe7_2", "scene_id": "a7631adbe7", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 25 cm away.", "target": "mug", "split": "validation"}
{"id": "a7631adbe7_3", "scene_id": "a7631adbe7", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 24.9}]}", "target": "mug", "split": "validation"}
{"id": "a7631adbe7_4", "scene_id": "a7631adbe7", "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": "validation"}
{"id": "a7631adbe7_5", "scene_id": "a7631adbe7", "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. washing machine\nC. coffee maker\nD. wet wipe", "choices": ["A. box", "B. washing machine", "C. coffee maker", "D. wet wipe"], "answer": "D. wet wipe", "target": "power strip", "split": "validation"}
{"id": "a7631adbe7_6", "scene_id": "a7631adbe7", "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 coffee table. It is also next to the wet wipe, about 6 cm away. It is further near the box, about 20 cm away.", "target": "power strip", "split": "validation"}
{"id": "a7631adbe7_7", "scene_id": "a7631adbe7", "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\": \"coffee table\", \"references\": [{\"object\": \"wet wipe\", \"relation\": \"next_to\", \"distance_cm\": 5.5}, {\"object\": \"box\", \"relation\": \"near\", \"distance_cm\": 20.4}]}", "target": "power strip", "split": "validation"}
{"id": "a766294249_0", "scene_id": "a766294249", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "a766294249_1", "scene_id": "a766294249", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. table\nC. curtain\nD. chair", "choices": ["A. laptop", "B. table", "C. curtain", "D. chair"], "answer": "B. table", "target": "backpack", "split": "validation"}
{"id": "a766294249_2", "scene_id": "a766294249", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 table, 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 chair, about 14 cm away.", "target": "backpack", "split": "validation"}
{"id": "a766294249_3", "scene_id": "a766294249", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"next_to\", \"distance_cm\": 4.9}, {\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 10.8}]}", "target": "backpack", "split": "validation"}
{"id": "a766294249_4", "scene_id": "a766294249", "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": "validation"}
{"id": "a766294249_5", "scene_id": "a766294249", "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. book\nB. basket\nC. books\nD. spray bottle", "choices": ["A. book", "B. basket", "C. books", "D. spray bottle"], "answer": "A. book", "target": "bottle", "split": "validation"}
{"id": "a766294249_6", "scene_id": "a766294249", "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 below the basket, about 6 cm away. It is further next to the books, about 7 cm away.", "target": "bottle", "split": "validation"}
{"id": "a766294249_7", "scene_id": "a766294249", "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\": \"basket\", \"relation\": \"side_below\", \"distance_cm\": 5.9}, {\"object\": \"books\", \"relation\": \"next_to\", \"distance_cm\": 7.4}]}", "target": "bottle", "split": "validation"}
{"id": "a766294249_8", "scene_id": "a766294249", "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": "validation"}
{"id": "a766294249_9", "scene_id": "a766294249", "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. book\nB. laptop\nC. head phones\nD. bottle", "choices": ["A. book", "B. laptop", "C. head phones", "D. bottle"], "answer": "B. laptop", "target": "spray bottle", "split": "validation"}
{"id": "a766294249_10", "scene_id": "a766294249", "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 laptop, about 6 cm away. It is also to the side of and below the bottle, about 8 cm away. It is further to the side of and below the book, about 16 cm away.", "target": "spray bottle", "split": "validation"}
{"id": "a766294249_11", "scene_id": "a766294249", "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\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 6.3}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 8.2}]}", "target": "spray bottle", "split": "validation"}
{"id": "a892730b61_0", "scene_id": "a892730b61", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "a892730b61_1", "scene_id": "a892730b61", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. wall lamp\nB. table\nC. keyboard\nD. monitor", "choices": ["A. wall lamp", "B. table", "C. keyboard", "D. monitor"], "answer": "A. wall lamp", "target": "cable", "split": "validation"}
{"id": "a892730b61_2", "scene_id": "a892730b61", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 wall lamp, about 53 cm away.", "target": "cable", "split": "validation"}
{"id": "a892730b61_3", "scene_id": "a892730b61", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"wall lamp\", \"references\": [{\"object\": \"wall lamp\", \"relation\": \"side_above\", \"distance_cm\": 53.0}]}", "target": "cable", "split": "validation"}
{"id": "ab6983ae6c_0", "scene_id": "ab6983ae6c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "ab6983ae6c_1", "scene_id": "ab6983ae6c", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. headphone\nB. bottle\nC. monitor\nD. table", "choices": ["A. headphone", "B. bottle", "C. monitor", "D. table"], "answer": "D. table", "target": "cable", "split": "validation"}
{"id": "ab6983ae6c_2", "scene_id": "ab6983ae6c", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 headphone, about 26 cm away. It is further to the side of and above the office chair, about 42 cm away.", "target": "cable", "split": "validation"}
{"id": "ab6983ae6c_3", "scene_id": "ab6983ae6c", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"headphone\", \"relation\": \"near\", \"distance_cm\": 26.4}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 41.6}]}", "target": "cable", "split": "validation"}
{"id": "ab6983ae6c_4", "scene_id": "ab6983ae6c", "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": "validation"}
{"id": "ab6983ae6c_5", "scene_id": "ab6983ae6c", "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. office chair\nC. monitor\nD. headphone", "choices": ["A. bottle", "B. office chair", "C. monitor", "D. headphone"], "answer": "B. office chair", "target": "cup", "split": "validation"}
{"id": "ab6983ae6c_6", "scene_id": "ab6983ae6c", "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 office chair, about 7 cm away. It is further to the side of and below the bottle, about 16 cm away.", "target": "cup", "split": "validation"}
{"id": "ab6983ae6c_7", "scene_id": "ab6983ae6c", "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\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 7.0}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 15.8}]}", "target": "cup", "split": "validation"}
{"id": "ab6983ae6c_8", "scene_id": "ab6983ae6c", "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": "validation"}
{"id": "ab6983ae6c_9", "scene_id": "ab6983ae6c", "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. computer tower\nB. keyboard\nC. mouse\nD. heater", "choices": ["A. computer tower", "B. keyboard", "C. mouse", "D. heater"], "answer": "A. computer tower", "target": "mug", "split": "validation"}
{"id": "ab6983ae6c_10", "scene_id": "ab6983ae6c", "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 under the computer tower, about 3 cm away. It is further near the mouse, about 16 cm away.", "target": "mug", "split": "validation"}
{"id": "ab6983ae6c_11", "scene_id": "ab6983ae6c", "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\": \"computer tower\", \"relation\": \"under\", \"distance_cm\": 2.6}, {\"object\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 15.6}]}", "target": "mug", "split": "validation"}
{"id": "ab6983ae6c_12", "scene_id": "ab6983ae6c", "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": "validation"}
{"id": "ab6983ae6c_13", "scene_id": "ab6983ae6c", "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. mug\nB. table\nC. outlet\nD. split", "choices": ["A. mug", "B. table", "C. outlet", "D. split"], "answer": "A. mug", "target": "pen holder", "split": "validation"}
{"id": "ab6983ae6c_14", "scene_id": "ab6983ae6c", "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 above the mug, about 14 cm away. It is also to the side of and above the table, about 16 cm away. It is further to the side of and above the outlet, about 24 cm away.", "target": "pen holder", "split": "validation"}
{"id": "ab6983ae6c_15", "scene_id": "ab6983ae6c", "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\": \"mug\", \"relation\": \"side_above\", \"distance_cm\": 14.2}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 15.6}]}", "target": "pen holder", "split": "validation"}
{"id": "ab6983ae6c_16", "scene_id": "ab6983ae6c", "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 usb hub appear?", "choices": null, "answer": "No", "target": "usb hub", "split": "validation"}
{"id": "ab6983ae6c_17", "scene_id": "ab6983ae6c", "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 in this scene?\nA. paper\nB. box\nC. bottle\nD. table", "choices": ["A. paper", "B. box", "C. bottle", "D. table"], "answer": "B. box", "target": "tissue", "split": "validation"}
{"id": "ab6983ae6c_18", "scene_id": "ab6983ae6c", "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 in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The tissue is on the cabinet. It is also next to the box, about 4 cm away. It is further next to the paper, about 10 cm away.", "target": "tissue", "split": "validation"}
{"id": "ab6983ae6c_19", "scene_id": "ab6983ae6c", "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 in structured JSON format.", "choices": null, "answer": "{\"target\": \"tissue\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 3.5}, {\"object\": \"paper\", \"relation\": \"next_to\", \"distance_cm\": 9.7}]}", "target": "tissue", "split": "validation"}
{"id": "ad562ce09b_0", "scene_id": "ad562ce09b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "ad562ce09b_1", "scene_id": "ad562ce09b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. headphones\nB. laptop\nC. power extender\nD. laptop charger", "choices": ["A. headphones", "B. laptop", "C. power extender", "D. laptop charger"], "answer": "C. power extender", "target": "backpack", "split": "validation"}
{"id": "ad562ce09b_2", "scene_id": "ad562ce09b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 extender, about 15 cm away. It is also to the side of and above the laptop charger, about 26 cm away. It is further to the side of and below the laptop, about 43 cm away.", "target": "backpack", "split": "validation"}
{"id": "ad562ce09b_3", "scene_id": "ad562ce09b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"power extender\", \"references\": [{\"object\": \"power extender\", \"relation\": \"side_above\", \"distance_cm\": 15.2}, {\"object\": \"laptop charger\", \"relation\": \"side_above\", \"distance_cm\": 25.5}]}", "target": "backpack", "split": "validation"}
{"id": "ad562ce09b_4", "scene_id": "ad562ce09b", "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": "validation"}
{"id": "ad562ce09b_5", "scene_id": "ad562ce09b", "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. vase\nC. books\nD. backpack", "choices": ["A. shelf", "B. vase", "C. books", "D. backpack"], "answer": "A. shelf", "target": "bottle", "split": "validation"}
{"id": "ad562ce09b_6", "scene_id": "ad562ce09b", "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 to the side of and below the books, about 18 cm away. It is further to the side of and above the vase, about 54 cm away.", "target": "bottle", "split": "validation"}
{"id": "ad562ce09b_7", "scene_id": "ad562ce09b", "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\": \"books\", \"relation\": \"side_below\", \"distance_cm\": 17.5}, {\"object\": \"vase\", \"relation\": \"side_above\", \"distance_cm\": 53.6}]}", "target": "bottle", "split": "validation"}
{"id": "ad562ce09b_8", "scene_id": "ad562ce09b", "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": "validation"}
{"id": "ad562ce09b_9", "scene_id": "ad562ce09b", "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. laptop\nB. laptop charger\nC. mouse\nD. chair", "choices": ["A. laptop", "B. laptop charger", "C. mouse", "D. chair"], "answer": "C. mouse", "target": "headphones", "split": "validation"}
{"id": "ad562ce09b_10", "scene_id": "ad562ce09b", "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 near the mouse, about 22 cm away. It is further to the side of and above the laptop charger, about 24 cm away.", "target": "headphones", "split": "validation"}
{"id": "ad562ce09b_11", "scene_id": "ad562ce09b", "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\": \"mouse\", \"relation\": \"near\", \"distance_cm\": 22.2}, {\"object\": \"laptop charger\", \"relation\": \"side_above\", \"distance_cm\": 23.9}]}", "target": "headphones", "split": "validation"}
{"id": "ad562ce09b_12", "scene_id": "ad562ce09b", "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 charger appear?", "choices": null, "answer": "No", "target": "charger", "split": "validation"}
{"id": "ad562ce09b_13", "scene_id": "ad562ce09b", "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. headphones\nB. laptop\nC. chair\nD. towel", "choices": ["A. headphones", "B. laptop", "C. chair", "D. towel"], "answer": "B. laptop", "target": "mouse", "split": "validation"}
{"id": "ad562ce09b_14", "scene_id": "ad562ce09b", "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 laptop, about 6 cm away. It is further near the headphones, about 22 cm away.", "target": "mouse", "split": "validation"}
{"id": "ad562ce09b_15", "scene_id": "ad562ce09b", "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\": \"next_to\", \"distance_cm\": 5.5}, {\"object\": \"headphones\", \"relation\": \"near\", \"distance_cm\": 22.2}]}", "target": "mouse", "split": "validation"}
{"id": "ad562ce09b_16", "scene_id": "ad562ce09b", "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 tv controller appear?", "choices": null, "answer": "No", "target": "tv controller", "split": "validation"}
{"id": "ad562ce09b_17", "scene_id": "ad562ce09b", "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 in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The phone is on the bed. It is also next to the laptop, about 10 cm away. It is further next to the notebook, about 10 cm away.", "target": "phone", "split": "validation"}
{"id": "ad562ce09b_18", "scene_id": "ad562ce09b", "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 in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone\", \"support_surface\": \"bed\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 9.6}, {\"object\": \"notebook\", \"relation\": \"next_to\", \"distance_cm\": 9.9}]}", "target": "phone", "split": "validation"}
{"id": "af112b8903_0", "scene_id": "af112b8903", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "af112b8903_1", "scene_id": "af112b8903", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. cubicle separator\nC. table\nD. monitor", "choices": ["A. cabinet", "B. cubicle separator", "C. table", "D. monitor"], "answer": "B. cubicle separator", "target": "bottle", "split": "validation"}
{"id": "af112b8903_2", "scene_id": "af112b8903", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 cubicle separator, about 5 cm away. It is further to the side of and above the cabinet, about 13 cm away.", "target": "bottle", "split": "validation"}
{"id": "af112b8903_3", "scene_id": "af112b8903", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"cubicle separator\", \"relation\": \"next_to\", \"distance_cm\": 4.8}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 13.1}]}", "target": "bottle", "split": "validation"}
{"id": "af112b8903_4", "scene_id": "af112b8903", "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": "validation"}
{"id": "af112b8903_5", "scene_id": "af112b8903", "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. cabinet\nB. rolling chair\nC. keyboard\nD. monitor", "choices": ["A. cabinet", "B. rolling chair", "C. keyboard", "D. monitor"], "answer": "C. keyboard", "target": "mouse", "split": "validation"}
{"id": "af112b8903_6", "scene_id": "af112b8903", "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 keyboard, about 17 cm away. It is further to the side of and above the cabinet, about 22 cm away.", "target": "mouse", "split": "validation"}
{"id": "af112b8903_7", "scene_id": "af112b8903", "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\": \"near\", \"distance_cm\": 17.2}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 21.9}]}", "target": "mouse", "split": "validation"}
{"id": "b0b004c40f_0", "scene_id": "b0b004c40f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "b0b004c40f_1", "scene_id": "b0b004c40f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. laptop\nD. duffel bag", "choices": ["A. chair", "B. table", "C. laptop", "D. duffel bag"], "answer": "D. duffel bag", "target": "backpack", "split": "validation"}
{"id": "b0b004c40f_2", "scene_id": "b0b004c40f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 duffel bag, about 11 cm away. It is further to the side of and above the laptop, about 13 cm away.", "target": "backpack", "split": "validation"}
{"id": "b0b004c40f_3", "scene_id": "b0b004c40f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"duffel bag\", \"relation\": \"next_to\", \"distance_cm\": 11.1}, {\"object\": \"laptop\", \"relation\": \"side_above\", \"distance_cm\": 12.6}]}", "target": "backpack", "split": "validation"}
{"id": "b2632b738a_0", "scene_id": "b2632b738a", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "b2632b738a_1", "scene_id": "b2632b738a", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. tube\nC. dresser\nD. light bulb", "choices": ["A. water bottle", "B. tube", "C. dresser", "D. light bulb"], "answer": "C. dresser", "target": "backpack", "split": "validation"}
{"id": "b2632b738a_2", "scene_id": "b2632b738a", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 dresser, about 8 cm away. It is further to the side of and below the water bottle, about 19 cm away.", "target": "backpack", "split": "validation"}
{"id": "b2632b738a_3", "scene_id": "b2632b738a", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"dresser\", \"relation\": \"side_above\", \"distance_cm\": 8.4}, {\"object\": \"water bottle\", \"relation\": \"side_below\", \"distance_cm\": 18.6}]}", "target": "backpack", "split": "validation"}
{"id": "b2632b738a_4", "scene_id": "b2632b738a", "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": "validation"}
{"id": "b2632b738a_5", "scene_id": "b2632b738a", "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. desk\nB. plant pot\nC. desk organiser\nD. curtain", "choices": ["A. desk", "B. plant pot", "C. desk organiser", "D. curtain"], "answer": "A. desk", "target": "bottle", "split": "validation"}
{"id": "b2632b738a_6", "scene_id": "b2632b738a", "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 desk, about 4 cm away. It is also near the desk organiser, about 16 cm away. It is further to the side of and above the plant pot, about 19 cm away.", "target": "bottle", "split": "validation"}
{"id": "b2632b738a_7", "scene_id": "b2632b738a", "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\": \"desk\", \"relation\": \"side_above\", \"distance_cm\": 4.3}, {\"object\": \"desk organiser\", \"relation\": \"near\", \"distance_cm\": 16.1}]}", "target": "bottle", "split": "validation"}
{"id": "b2632b738a_8", "scene_id": "b2632b738a", "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": "Yes", "target": "charger", "split": "validation"}
{"id": "b2632b738a_9", "scene_id": "b2632b738a", "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 charger in this scene?\nA. curtain\nB. table lamp\nC. desk organiser\nD. desk lamp", "choices": ["A. curtain", "B. table lamp", "C. desk organiser", "D. desk lamp"], "answer": "B. table lamp", "target": "charger", "split": "validation"}
{"id": "b2632b738a_10", "scene_id": "b2632b738a", "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 charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The charger is on the desk. It is also next to the table lamp, about 11 cm away. It is further to the side of and below the curtain, about 19 cm away.", "target": "charger", "split": "validation"}
{"id": "b2632b738a_11", "scene_id": "b2632b738a", "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 charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"charger\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"table lamp\", \"relation\": \"next_to\", \"distance_cm\": 10.7}, {\"object\": \"curtain\", \"relation\": \"side_below\", \"distance_cm\": 19.1}]}", "target": "charger", "split": "validation"}
{"id": "b2632b738a_12", "scene_id": "b2632b738a", "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": "validation"}
{"id": "b2632b738a_13", "scene_id": "b2632b738a", "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. files\nB. bookshelf\nC. wheeled chair\nD. mirror", "choices": ["A. files", "B. bookshelf", "C. wheeled chair", "D. mirror"], "answer": "C. wheeled chair", "target": "mouse", "split": "validation"}
{"id": "b2632b738a_14", "scene_id": "b2632b738a", "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 to the side of and above the wheeled chair, about 13 cm away. It is also to the side of and below the files, about 17 cm away. It is further to the side of and above the bookshelf, about 26 cm away.", "target": "mouse", "split": "validation"}
{"id": "b2632b738a_15", "scene_id": "b2632b738a", "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\": \"cabinet\", \"references\": [{\"object\": \"wheeled chair\", \"relation\": \"side_above\", \"distance_cm\": 13.4}, {\"object\": \"files\", \"relation\": \"side_below\", \"distance_cm\": 16.6}]}", "target": "mouse", "split": "validation"}
{"id": "b2632b738a_16", "scene_id": "b2632b738a", "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 scissors appear?", "choices": null, "answer": "No", "target": "scissors", "split": "validation"}
{"id": "b2632b738a_17", "scene_id": "b2632b738a", "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. tin box\nB. wall lamp\nC. umbrella\nD. shoes", "choices": ["A. tin box", "B. wall lamp", "C. umbrella", "D. shoes"], "answer": "A. tin box", "target": "mug", "split": "validation"}
{"id": "b2632b738a_18", "scene_id": "b2632b738a", "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 cabinet. It is also next to the tin box, about 3 cm away. It is further to the side of and above the umbrella, about 32 cm away.", "target": "mug", "split": "validation"}
{"id": "b2632b738a_19", "scene_id": "b2632b738a", "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\": \"tin box\", \"relation\": \"next_to\", \"distance_cm\": 2.8}, {\"object\": \"umbrella\", \"relation\": \"side_above\", \"distance_cm\": 31.8}]}", "target": "mug", "split": "validation"}
{"id": "b2632b738a_20", "scene_id": "b2632b738a", "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": "validation"}
{"id": "b2632b738a_21", "scene_id": "b2632b738a", "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. tube\nB. backpack\nC. microwave\nD. light bulb", "choices": ["A. tube", "B. backpack", "C. microwave", "D. light bulb"], "answer": "A. tube", "target": "water bottle", "split": "validation"}
{"id": "b2632b738a_22", "scene_id": "b2632b738a", "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 on the bookshelf. It is also to the side of and above the tube, about 7 cm away. It is further to the side of and above the microwave, about 17 cm away.", "target": "water bottle", "split": "validation"}
{"id": "b2632b738a_23", "scene_id": "b2632b738a", "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\": \"bookshelf\", \"references\": [{\"object\": \"tube\", \"relation\": \"side_above\", \"distance_cm\": 6.8}, {\"object\": \"microwave\", \"relation\": \"side_above\", \"distance_cm\": 16.8}]}", "target": "water bottle", "split": "validation"}
{"id": "b26e64c4b0_0", "scene_id": "b26e64c4b0", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "b26e64c4b0_1", "scene_id": "b26e64c4b0", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. blinds\nC. shoe\nD. chair", "choices": ["A. box", "B. blinds", "C. shoe", "D. chair"], "answer": "C. shoe", "target": "backpack", "split": "validation"}
{"id": "b26e64c4b0_2", "scene_id": "b26e64c4b0", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 shoe, about 9 cm away. It is also to the side of and below the box, about 42 cm away.", "target": "backpack", "split": "validation"}
{"id": "b26e64c4b0_3", "scene_id": "b26e64c4b0", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"shoe\", \"relation\": \"side_above\", \"distance_cm\": 9.2}, {\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 42.0}]}", "target": "backpack", "split": "validation"}
{"id": "b26e64c4b0_4", "scene_id": "b26e64c4b0", "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": "No", "target": "lanyard", "split": "validation"}
{"id": "b26e64c4b0_5", "scene_id": "b26e64c4b0", "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. guitar\nB. monitor\nC. blinds\nD. mouse", "choices": ["A. guitar", "B. monitor", "C. blinds", "D. mouse"], "answer": "A. guitar", "target": "cup", "split": "validation"}
{"id": "b26e64c4b0_6", "scene_id": "b26e64c4b0", "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 guitar, about 11 cm away. It is further next to the mouse, about 13 cm away.", "target": "cup", "split": "validation"}
{"id": "b26e64c4b0_7", "scene_id": "b26e64c4b0", "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\": \"guitar\", \"relation\": \"side_below\", \"distance_cm\": 10.7}, {\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 13.4}]}", "target": "cup", "split": "validation"}
{"id": "b26e64c4b0_8", "scene_id": "b26e64c4b0", "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": "validation"}
{"id": "b26e64c4b0_9", "scene_id": "b26e64c4b0", "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. plant\nB. mouse\nC. plant pot\nD. blinds", "choices": ["A. plant", "B. mouse", "C. plant pot", "D. blinds"], "answer": "B. mouse", "target": "keyboard", "split": "validation"}
{"id": "b26e64c4b0_10", "scene_id": "b26e64c4b0", "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 to the side of and below the blinds, about 8 cm away.", "target": "keyboard", "split": "validation"}
{"id": "b26e64c4b0_11", "scene_id": "b26e64c4b0", "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\": 5.0}, {\"object\": \"blinds\", \"relation\": \"side_below\", \"distance_cm\": 7.7}]}", "target": "keyboard", "split": "validation"}
{"id": "b26e64c4b0_12", "scene_id": "b26e64c4b0", "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 appear?", "choices": null, "answer": "No", "target": "pencil", "split": "validation"}
{"id": "b26e64c4b0_13", "scene_id": "b26e64c4b0", "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 monitor, about 4 cm away. It is also next to the keyboard, about 5 cm away. It is further to the side of and above the table, about 6 cm away.", "target": "mouse", "split": "validation"}
{"id": "b26e64c4b0_14", "scene_id": "b26e64c4b0", "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\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 4.0}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 5.0}]}", "target": "mouse", "split": "validation"}
{"id": "b26e64c4b0_15", "scene_id": "b26e64c4b0", "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 eyeglasses appear?", "choices": null, "answer": "No", "target": "eyeglasses", "split": "validation"}
{"id": "b26e64c4b0_16", "scene_id": "b26e64c4b0", "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 pad in this scene?\nA. chair\nB. pc\nC. guitar\nD. monitor", "choices": ["A. chair", "B. pc", "C. guitar", "D. monitor"], "answer": "D. monitor", "target": "mouse pad", "split": "validation"}
{"id": "b26e64c4b0_17", "scene_id": "b26e64c4b0", "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 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 6 cm away. It is further to the side of and below the guitar, about 10 cm away.", "target": "mouse pad", "split": "validation"}
{"id": "b26e64c4b0_18", "scene_id": "b26e64c4b0", "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 pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 6.1}, {\"object\": \"guitar\", \"relation\": \"side_below\", \"distance_cm\": 9.9}]}", "target": "mouse pad", "split": "validation"}
{"id": "b37c033757_0", "scene_id": "b37c033757", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "b37c033757_1", "scene_id": "b37c033757", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. rolling cart\nB. cable\nC. vacuum cleaner\nD. barrel", "choices": ["A. rolling cart", "B. cable", "C. vacuum cleaner", "D. barrel"], "answer": "D. barrel", "target": "backpack", "split": "validation"}
{"id": "b37c033757_2", "scene_id": "b37c033757", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 barrel, about 3 cm away. It is also to the side of and below the rolling cart, about 6 cm away. It is further to the side of and below the vacuum cleaner, about 23 cm away.", "target": "backpack", "split": "validation"}
{"id": "b37c033757_3", "scene_id": "b37c033757", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"barrel\", \"references\": [{\"object\": \"barrel\", \"relation\": \"under\", \"distance_cm\": 3.1}, {\"object\": \"rolling cart\", \"relation\": \"side_below\", \"distance_cm\": 6.4}]}", "target": "backpack", "split": "validation"}
{"id": "b8d441d736_0", "scene_id": "b8d441d736", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "b8d441d736_1", "scene_id": "b8d441d736", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. headphone case\nB. clipboard\nC. duffel bag\nD. chair", "choices": ["A. headphone case", "B. clipboard", "C. duffel bag", "D. chair"], "answer": "D. chair", "target": "backpack", "split": "validation"}
{"id": "b8d441d736_2", "scene_id": "b8d441d736", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 19 cm away. It is further near the headphone case, about 27 cm away.", "target": "backpack", "split": "validation"}
{"id": "b8d441d736_3", "scene_id": "b8d441d736", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 19.2}, {\"object\": \"headphone case\", \"relation\": \"near\", \"distance_cm\": 26.9}]}", "target": "backpack", "split": "validation"}
{"id": "b8d441d736_4", "scene_id": "b8d441d736", "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": "validation"}
{"id": "b8d441d736_5", "scene_id": "b8d441d736", "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. chair\nB. duffel bag\nC. backpack\nD. clipboard", "choices": ["A. chair", "B. duffel bag", "C. backpack", "D. clipboard"], "answer": "A. chair", "target": "headphone case", "split": "validation"}
{"id": "b8d441d736_6", "scene_id": "b8d441d736", "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 on the table. It is also to the side of and above the chair, about 16 cm away. It is further near the backpack, about 27 cm away.", "target": "headphone case", "split": "validation"}
{"id": "b8d441d736_7", "scene_id": "b8d441d736", "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\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 16.1}, {\"object\": \"backpack\", \"relation\": \"near\", \"distance_cm\": 26.9}]}", "target": "headphone case", "split": "validation"}
{"id": "bb05a0c48c_0", "scene_id": "bb05a0c48c", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "bb05a0c48c_1", "scene_id": "bb05a0c48c", "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 box, about 21 cm away. It is also near the tube, about 23 cm away. It is further to the side of and above the file folder, about 26 cm away.", "target": "bottle", "split": "validation"}
{"id": "bb05a0c48c_2", "scene_id": "bb05a0c48c", "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\": \"book shelf\", \"references\": [{\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 21.0}, {\"object\": \"tube\", \"relation\": \"near\", \"distance_cm\": 22.6}]}", "target": "bottle", "split": "validation"}
{"id": "bb05a0c48c_3", "scene_id": "bb05a0c48c", "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 lotion bottle appear?", "choices": null, "answer": "No", "target": "lotion bottle", "split": "validation"}
{"id": "bb05a0c48c_4", "scene_id": "bb05a0c48c", "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 strip in this scene?\nA. bedside table\nB. mop\nC. bed\nD. cleaning bucket", "choices": ["A. bedside table", "B. mop", "C. bed", "D. cleaning bucket"], "answer": "A. bedside table", "target": "power strip", "split": "validation"}
{"id": "bb05a0c48c_5", "scene_id": "bb05a0c48c", "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 strip in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The power strip is under the bedside table. It is also to the side of and below the cleaning bucket, about 24 cm away. It is further to the side of and below the bed, about 32 cm away.", "target": "power strip", "split": "validation"}
{"id": "bb05a0c48c_6", "scene_id": "bb05a0c48c", "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 strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"bedside table\", \"references\": [{\"object\": \"bedside table\", \"relation\": \"under\", \"distance_cm\": 2.3}, {\"object\": \"cleaning bucket\", \"relation\": \"side_below\", \"distance_cm\": 24.0}]}", "target": "power strip", "split": "validation"}
{"id": "bc400d86e1_0", "scene_id": "bc400d86e1", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "bc400d86e1_1", "scene_id": "bc400d86e1", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. computer tower\nB. cd player\nC. mouse\nD. mouse pad", "choices": ["A. computer tower", "B. cd player", "C. mouse", "D. mouse pad"], "answer": "D. mouse pad", "target": "headset", "split": "validation"}
{"id": "bc400d86e1_2", "scene_id": "bc400d86e1", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 mouse pad, about 5 cm away. It is also to the side of and below the computer tower, about 6 cm away. It is further next to the cd player, about 7 cm away.", "target": "headset", "split": "validation"}
{"id": "bc400d86e1_3", "scene_id": "bc400d86e1", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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 table\", \"references\": [{\"object\": \"mouse pad\", \"relation\": \"next_to\", \"distance_cm\": 4.5}, {\"object\": \"computer tower\", \"relation\": \"side_below\", \"distance_cm\": 6.1}]}", "target": "headset", "split": "validation"}
{"id": "bc400d86e1_4", "scene_id": "bc400d86e1", "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": "validation"}
{"id": "bc400d86e1_5", "scene_id": "bc400d86e1", "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 pad in this scene?\nA. keyboard\nB. headset\nC. monitor\nD. cd player", "choices": ["A. keyboard", "B. headset", "C. monitor", "D. cd player"], "answer": "B. headset", "target": "mouse pad", "split": "validation"}
{"id": "bc400d86e1_6", "scene_id": "bc400d86e1", "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 pad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mouse pad is next to the headset, about 5 cm away. It is also to the side of and below the cd player, about 6 cm away. It is further next to the keyboard, about 12 cm away.", "target": "mouse pad", "split": "validation"}
{"id": "bc400d86e1_7", "scene_id": "bc400d86e1", "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 pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"office table\", \"references\": [{\"object\": \"headset\", \"relation\": \"next_to\", \"distance_cm\": 4.5}, {\"object\": \"cd player\", \"relation\": \"side_below\", \"distance_cm\": 6.4}]}", "target": "mouse pad", "split": "validation"}
{"id": "bc400d86e1_8", "scene_id": "bc400d86e1", "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": "validation"}
{"id": "bc400d86e1_9", "scene_id": "bc400d86e1", "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. box\nB. books\nC. storage cabinet\nD. folders", "choices": ["A. box", "B. books", "C. storage cabinet", "D. folders"], "answer": "A. box", "target": "stapler", "split": "validation"}
{"id": "bc400d86e1_10", "scene_id": "bc400d86e1", "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 box, about 20 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 storage cabinet, about 27 cm away.", "target": "stapler", "split": "validation"}
{"id": "bc400d86e1_11", "scene_id": "bc400d86e1", "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\": \"storage rack\", \"references\": [{\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 20.1}, {\"object\": \"books\", \"relation\": \"side_below\", \"distance_cm\": 26.4}]}", "target": "stapler", "split": "validation"}
{"id": "bcbc6e1d2d_0", "scene_id": "bcbc6e1d2d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "bcbc6e1d2d_1", "scene_id": "bcbc6e1d2d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. chair\nC. notebook\nD. whiteboard", "choices": ["A. table", "B. chair", "C. notebook", "D. whiteboard"], "answer": "A. table", "target": "bottle", "split": "validation"}
{"id": "bcbc6e1d2d_2", "scene_id": "bcbc6e1d2d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 5 cm away. It is also to the side of and above the chair, about 37 cm away. It is further to the side of and below the whiteboard, about 91 cm away.", "target": "bottle", "split": "validation"}
{"id": "bcbc6e1d2d_3", "scene_id": "bcbc6e1d2d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 4.9}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 37.0}]}", "target": "bottle", "split": "validation"}
{"id": "bcbc6e1d2d_4", "scene_id": "bcbc6e1d2d", "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": "validation"}
{"id": "bcbc6e1d2d_5", "scene_id": "bcbc6e1d2d", "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. chair\nB. bottle\nC. table\nD. paper napkin", "choices": ["A. chair", "B. bottle", "C. table", "D. paper napkin"], "answer": "A. chair", "target": "notebook", "split": "validation"}
{"id": "bcbc6e1d2d_6", "scene_id": "bcbc6e1d2d", "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 to the side of and above the chair, about 28 cm away.", "target": "notebook", "split": "validation"}
{"id": "bcbc6e1d2d_7", "scene_id": "bcbc6e1d2d", "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\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 27.9}]}", "target": "notebook", "split": "validation"}
{"id": "bd7375297e_0", "scene_id": "bd7375297e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "bd7375297e_1", "scene_id": "bd7375297e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. power adapter\nB. electrical outlet\nC. bottle\nD. paper", "choices": ["A. power adapter", "B. electrical outlet", "C. bottle", "D. paper"], "answer": "A. power adapter", "target": "pencil case", "split": "validation"}
{"id": "bd7375297e_2", "scene_id": "bd7375297e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 power adapter, about 8 cm away. It is further to the side of and below the bottle, about 13 cm away.", "target": "pencil case", "split": "validation"}
{"id": "bd7375297e_3", "scene_id": "bd7375297e", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"power adapter\", \"relation\": \"next_to\", \"distance_cm\": 8.2}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 12.8}]}", "target": "pencil case", "split": "validation"}
{"id": "bd7375297e_4", "scene_id": "bd7375297e", "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": "No", "target": "calculator", "split": "validation"}
{"id": "bd7375297e_5", "scene_id": "bd7375297e", "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. power socket\nB. soap dispenser\nC. object/toiletry\nD. sink", "choices": ["A. power socket", "B. soap dispenser", "C. object/toiletry", "D. sink"], "answer": "D. sink", "target": "toothbrush", "split": "validation"}
{"id": "bd7375297e_6", "scene_id": "bd7375297e", "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 to the side of and above the sink, about 3 cm away. It is also under the soap dispenser, about 5 cm away. It is further next to the object/toiletry, about 7 cm away.", "target": "toothbrush", "split": "validation"}
{"id": "bd7375297e_7", "scene_id": "bd7375297e", "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\": \"bathroom cabinet/bathroom counter\", \"references\": [{\"object\": \"sink\", \"relation\": \"side_above\", \"distance_cm\": 3.2}, {\"object\": \"soap dispenser\", \"relation\": \"under\", \"distance_cm\": 4.8}]}", "target": "toothbrush", "split": "validation"}
{"id": "bd7375297e_8", "scene_id": "bd7375297e", "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": "Yes", "target": "tv remote", "split": "validation"}
{"id": "bd7375297e_9", "scene_id": "bd7375297e", "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. box\nB. tv board\nC. pencil case\nD. bottle", "choices": ["A. box", "B. tv board", "C. pencil case", "D. bottle"], "answer": "A. box", "target": "tv remote", "split": "validation"}
{"id": "bd7375297e_10", "scene_id": "bd7375297e", "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 on the table. It is also to the side of and below the box, about 6 cm away. It is further to the side of and below the bottle, about 13 cm away.", "target": "tv remote", "split": "validation"}
{"id": "bd7375297e_11", "scene_id": "bd7375297e", "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\": \"table\", \"references\": [{\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 5.9}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 13.5}]}", "target": "tv remote", "split": "validation"}
{"id": "bd9305480d_0", "scene_id": "bd9305480d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "bd9305480d_1", "scene_id": "bd9305480d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. locker\nC. bullettin boards\nD. box", "choices": ["A. paper", "B. locker", "C. bullettin boards", "D. box"], "answer": "A. paper", "target": "bottle", "split": "validation"}
{"id": "bd9305480d_2", "scene_id": "bd9305480d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 6 cm away. It is further to the side of and above the locker, about 38 cm away.", "target": "bottle", "split": "validation"}
{"id": "bd9305480d_3", "scene_id": "bd9305480d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 6.3}, {\"object\": \"locker\", \"relation\": \"side_above\", \"distance_cm\": 38.3}]}", "target": "bottle", "split": "validation"}
{"id": "c06a983e63_0", "scene_id": "c06a983e63", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "c06a983e63_1", "scene_id": "c06a983e63", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. lamp\nB. monitor\nC. book\nD. box", "choices": ["A. lamp", "B. monitor", "C. book", "D. box"], "answer": "A. lamp", "target": "bottle", "split": "validation"}
{"id": "c06a983e63_2", "scene_id": "c06a983e63", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 under the lamp, about 5 cm away. It is further to the side of and below the monitor, about 22 cm away.", "target": "bottle", "split": "validation"}
{"id": "c06a983e63_3", "scene_id": "c06a983e63", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"lamp\", \"relation\": \"under\", \"distance_cm\": 4.6}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 21.8}]}", "target": "bottle", "split": "validation"}
{"id": "c06a983e63_4", "scene_id": "c06a983e63", "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": "validation"}
{"id": "c06a983e63_5", "scene_id": "c06a983e63", "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. folders\nB. office chair\nC. cabinet\nD. mouse", "choices": ["A. folders", "B. office chair", "C. cabinet", "D. mouse"], "answer": "D. mouse", "target": "keyboard", "split": "validation"}
{"id": "c06a983e63_6", "scene_id": "c06a983e63", "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 8 cm away. It is further near the folders, about 33 cm away.", "target": "keyboard", "split": "validation"}
{"id": "c06a983e63_7", "scene_id": "c06a983e63", "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\": 7.7}, {\"object\": \"folders\", \"relation\": \"near\", \"distance_cm\": 32.8}]}", "target": "keyboard", "split": "validation"}
{"id": "c06a983e63_8", "scene_id": "c06a983e63", "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": "validation"}
{"id": "c06a983e63_9", "scene_id": "c06a983e63", "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. cabinet\nC. keyboard\nD. phone", "choices": ["A. office chair", "B. cabinet", "C. keyboard", "D. phone"], "answer": "C. keyboard", "target": "mouse", "split": "validation"}
{"id": "c06a983e63_10", "scene_id": "c06a983e63", "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 above the cabinet, about 27 cm away.", "target": "mouse", "split": "validation"}
{"id": "c06a983e63_11", "scene_id": "c06a983e63", "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\": 7.7}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 26.8}]}", "target": "mouse", "split": "validation"}
{"id": "c06a983e63_12", "scene_id": "c06a983e63", "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": "validation"}
{"id": "c06a983e63_13", "scene_id": "c06a983e63", "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 in this scene?\nA. book\nB. box\nC. lamp\nD. monitor", "choices": ["A. book", "B. box", "C. lamp", "D. monitor"], "answer": "C. lamp", "target": "phone", "split": "validation"}
{"id": "c06a983e63_14", "scene_id": "c06a983e63", "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 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 below the lamp, about 6 cm away. It is further next to the box, about 15 cm away.", "target": "phone", "split": "validation"}
{"id": "c06a983e63_15", "scene_id": "c06a983e63", "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 in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone\", \"support_surface\": \"table\", \"references\": [{\"object\": \"lamp\", \"relation\": \"side_below\", \"distance_cm\": 6.2}, {\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 14.9}]}", "target": "phone", "split": "validation"}
{"id": "c07c8a276d_0", "scene_id": "c07c8a276d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "c07c8a276d_1", "scene_id": "c07c8a276d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. standing lamp\nB. cable strip\nC. power socket\nD. bed", "choices": ["A. standing lamp", "B. cable strip", "C. power socket", "D. bed"], "answer": "B. cable strip", "target": "backpack", "split": "validation"}
{"id": "c07c8a276d_2", "scene_id": "c07c8a276d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 cable strip, about 10 cm away. It is also to the side of and below the bed, about 30 cm away. It is further to the side of and below the standing lamp, about 37 cm away.", "target": "backpack", "split": "validation"}
{"id": "c07c8a276d_3", "scene_id": "c07c8a276d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"cable strip\", \"relation\": \"side_above\", \"distance_cm\": 9.7}, {\"object\": \"bed\", \"relation\": \"side_below\", \"distance_cm\": 29.7}]}", "target": "backpack", "split": "validation"}
{"id": "c07c8a276d_4", "scene_id": "c07c8a276d", "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": "validation"}
{"id": "c07c8a276d_5", "scene_id": "c07c8a276d", "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. kitchen paper roll\nC. standing lamp\nD. box", "choices": ["A. blanket", "B. kitchen paper roll", "C. standing lamp", "D. box"], "answer": "D. box", "target": "bottle", "split": "validation"}
{"id": "c07c8a276d_6", "scene_id": "c07c8a276d", "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 box, about 3 cm away. It is further next to the kitchen paper roll, about 6 cm away.", "target": "bottle", "split": "validation"}
{"id": "c07c8a276d_7", "scene_id": "c07c8a276d", "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\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 2.7}, {\"object\": \"kitchen paper roll\", \"relation\": \"next_to\", \"distance_cm\": 5.6}]}", "target": "bottle", "split": "validation"}
{"id": "c0c863b72d_0", "scene_id": "c0c863b72d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "c0c863b72d_1", "scene_id": "c0c863b72d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. chair\nC. cabinet\nD. table", "choices": ["A. paper", "B. chair", "C. cabinet", "D. table"], "answer": "D. table", "target": "bottle", "split": "validation"}
{"id": "c0c863b72d_2", "scene_id": "c0c863b72d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 paper, about 35 cm away. It is further to the side of and above the cabinet, about 39 cm away.", "target": "bottle", "split": "validation"}
{"id": "c0c863b72d_3", "scene_id": "c0c863b72d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 35.2}]}", "target": "bottle", "split": "validation"}
{"id": "c0c863b72d_4", "scene_id": "c0c863b72d", "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": "validation"}
{"id": "c0c863b72d_5", "scene_id": "c0c863b72d", "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 pad in this scene?\nA. cabinet\nB. office chair\nC. keyboard\nD. monitor", "choices": ["A. cabinet", "B. office chair", "C. keyboard", "D. monitor"], "answer": "C. keyboard", "target": "mouse pad", "split": "validation"}
{"id": "c0c863b72d_6", "scene_id": "c0c863b72d", "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 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 keyboard, about 8 cm away. It is further to the side of and below the monitor, about 14 cm away.", "target": "mouse pad", "split": "validation"}
{"id": "c0c863b72d_7", "scene_id": "c0c863b72d", "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 pad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse pad\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 8.3}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 13.6}]}", "target": "mouse pad", "split": "validation"}
{"id": "c0c863b72d_8", "scene_id": "c0c863b72d", "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 appear?", "choices": null, "answer": "No", "target": "phone", "split": "validation"}
{"id": "c0c863b72d_9", "scene_id": "c0c863b72d", "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, about 3 cm away. It is also to the side of and above the paper, about 3 cm away. It is further next to the box, about 4 cm away.", "target": "mug", "split": "validation"}
{"id": "c0c863b72d_10", "scene_id": "c0c863b72d", "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\": 2.6}, {\"object\": \"paper\", \"relation\": \"side_above\", \"distance_cm\": 3.3}]}", "target": "mug", "split": "validation"}
{"id": "c285c82ade_0", "scene_id": "c285c82ade", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "c285c82ade_1", "scene_id": "c285c82ade", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. papers\nB. light switch\nC. doorframe\nD. window door", "choices": ["A. papers", "B. light switch", "C. doorframe", "D. window door"], "answer": "D. window door", "target": "phone charger", "split": "validation"}
{"id": "c285c82ade_2", "scene_id": "c285c82ade", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 window door, about 29 cm away. It is further below the light switch, about 73 cm away.", "target": "phone charger", "split": "validation"}
{"id": "c285c82ade_3", "scene_id": "c285c82ade", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"window door\", \"relation\": \"side_below\", \"distance_cm\": 29.1}]}", "target": "phone charger", "split": "validation"}
{"id": "c285c82ade_4", "scene_id": "c285c82ade", "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": "validation"}
{"id": "c285c82ade_5", "scene_id": "c285c82ade", "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. doorframe\nB. window door\nC. papers\nD. light switch", "choices": ["A. doorframe", "B. window door", "C. papers", "D. light switch"], "answer": "A. doorframe", "target": "water bottle", "split": "validation"}
{"id": "c285c82ade_6", "scene_id": "c285c82ade", "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 to the side of and below the doorframe, about 41 cm away. It is also to the side of and above the door, about 42 cm away. It is further to the side of and above the window door, about 65 cm away.", "target": "water bottle", "split": "validation"}
{"id": "c285c82ade_7", "scene_id": "c285c82ade", "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\": \"door\", \"references\": [{\"object\": \"doorframe\", \"relation\": \"side_below\", \"distance_cm\": 40.9}, {\"object\": \"door\", \"relation\": \"side_above\", \"distance_cm\": 41.8}]}", "target": "water bottle", "split": "validation"}
{"id": "c2d714d386_0", "scene_id": "c2d714d386", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "c2d714d386_1", "scene_id": "c2d714d386", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. closet\nB. bookshelf\nC. table\nD. monitor", "choices": ["A. closet", "B. bookshelf", "C. table", "D. monitor"], "answer": "B. bookshelf", "target": "handbag", "split": "validation"}
{"id": "c2d714d386_2", "scene_id": "c2d714d386", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 bookshelf, about 27 cm away.", "target": "handbag", "split": "validation"}
{"id": "c2d714d386_3", "scene_id": "c2d714d386", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"bookshelf\", \"references\": [{\"object\": \"bookshelf\", \"relation\": \"side_below\", \"distance_cm\": 27.3}]}", "target": "handbag", "split": "validation"}
{"id": "c413b34238_0", "scene_id": "c413b34238", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "c413b34238_1", "scene_id": "c413b34238", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. decoration\nB. plant pot\nC. whiteboard\nD. plant", "choices": ["A. decoration", "B. plant pot", "C. whiteboard", "D. plant"], "answer": "D. plant", "target": "bottle", "split": "validation"}
{"id": "c413b34238_2", "scene_id": "c413b34238", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 under the plant, about 4 cm away. It is further next to the plant pot, about 7 cm away.", "target": "bottle", "split": "validation"}
{"id": "c413b34238_3", "scene_id": "c413b34238", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"plant\", \"relation\": \"under\", \"distance_cm\": 4.0}, {\"object\": \"plant pot\", \"relation\": \"next_to\", \"distance_cm\": 7.3}]}", "target": "bottle", "split": "validation"}
{"id": "c413b34238_4", "scene_id": "c413b34238", "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": "validation"}
{"id": "c413b34238_5", "scene_id": "c413b34238", "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. ring light\nD. monitor", "choices": ["A. keyboard", "B. chair", "C. ring light", "D. monitor"], "answer": "A. keyboard", "target": "mouse", "split": "validation"}
{"id": "c413b34238_6", "scene_id": "c413b34238", "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 to the side of and below the monitor, about 7 cm away. It is further to the side of and above the chair, about 18 cm away.", "target": "mouse", "split": "validation"}
{"id": "c413b34238_7", "scene_id": "c413b34238", "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\": \"chair\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 5.3}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 7.5}]}", "target": "mouse", "split": "validation"}
{"id": "c465f388d1_0", "scene_id": "c465f388d1", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "c465f388d1_1", "scene_id": "c465f388d1", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. beverage bottle\nB. mattress\nC. bedside table\nD. pillow", "choices": ["A. beverage bottle", "B. mattress", "C. bedside table", "D. pillow"], "answer": "A. beverage bottle", "target": "backpack", "split": "validation"}
{"id": "c465f388d1_2", "scene_id": "c465f388d1", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 above the beverage bottle, about 3 cm away. It is further to the side of and below the mattress, about 6 cm away.", "target": "backpack", "split": "validation"}
{"id": "c465f388d1_3", "scene_id": "c465f388d1", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"beverage bottle\", \"relation\": \"side_above\", \"distance_cm\": 3.4}, {\"object\": \"mattress\", \"relation\": \"side_below\", \"distance_cm\": 6.4}]}", "target": "backpack", "split": "validation"}
{"id": "c465f388d1_4", "scene_id": "c465f388d1", "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": "validation"}
{"id": "c465f388d1_5", "scene_id": "c465f388d1", "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. laptop\nB. box\nC. mouse\nD. book", "choices": ["A. laptop", "B. box", "C. mouse", "D. book"], "answer": "A. laptop", "target": "mug", "split": "validation"}
{"id": "c465f388d1_6", "scene_id": "c465f388d1", "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 laptop, about 7 cm away. It is further next to the box, about 13 cm away.", "target": "mug", "split": "validation"}
{"id": "c465f388d1_7", "scene_id": "c465f388d1", "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\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 7.0}, {\"object\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 12.6}]}", "target": "mug", "split": "validation"}
{"id": "c465f388d1_8", "scene_id": "c465f388d1", "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": "validation"}
{"id": "c465f388d1_9", "scene_id": "c465f388d1", "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. air freshener\nB. paper bag\nC. book\nD. pillow", "choices": ["A. air freshener", "B. paper bag", "C. book", "D. pillow"], "answer": "A. air freshener", "target": "phone charger", "split": "validation"}
{"id": "c465f388d1_10", "scene_id": "c465f388d1", "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 air freshener, about 12 cm away. It is also to the side of and below the paper bag, about 29 cm away. It is further near the pillow, about 35 cm away.", "target": "phone charger", "split": "validation"}
{"id": "c465f388d1_11", "scene_id": "c465f388d1", "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\": \"pillow\", \"references\": [{\"object\": \"air freshener\", \"relation\": \"side_above\", \"distance_cm\": 12.2}, {\"object\": \"paper bag\", \"relation\": \"side_below\", \"distance_cm\": 28.8}]}", "target": "phone charger", "split": "validation"}
{"id": "c856c41c99_0", "scene_id": "c856c41c99", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "c856c41c99_1", "scene_id": "c856c41c99", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. board\nB. dumbbell case\nC. monitor\nD. dumbbell", "choices": ["A. board", "B. dumbbell case", "C. monitor", "D. dumbbell"], "answer": "D. dumbbell", "target": "spray bottle", "split": "validation"}
{"id": "c856c41c99_2", "scene_id": "c856c41c99", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 dumbbell, about 10 cm away. It is also to the side of and below the dumbbell case, about 56 cm away.", "target": "spray bottle", "split": "validation"}
{"id": "c856c41c99_3", "scene_id": "c856c41c99", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"dumbbell case\", \"references\": [{\"object\": \"dumbbell\", \"relation\": \"next_to\", \"distance_cm\": 9.9}, {\"object\": \"dumbbell case\", \"relation\": \"side_below\", \"distance_cm\": 56.4}]}", "target": "spray bottle", "split": "validation"}
{"id": "c8d099ecd8_0", "scene_id": "c8d099ecd8", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "c8d099ecd8_1", "scene_id": "c8d099ecd8", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. desk\nC. extension board\nD. heater", "choices": ["A. office chair", "B. desk", "C. extension board", "D. heater"], "answer": "A. office chair", "target": "extension cord", "split": "validation"}
{"id": "c8d099ecd8_2", "scene_id": "c8d099ecd8", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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. It is also to the side of and above the office chair, about 12 cm away.", "target": "extension cord", "split": "validation"}
{"id": "c8d099ecd8_3", "scene_id": "c8d099ecd8", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 12.3}]}", "target": "extension cord", "split": "validation"}
{"id": "c8d099ecd8_4", "scene_id": "c8d099ecd8", "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": "validation"}
{"id": "c8d099ecd8_5", "scene_id": "c8d099ecd8", "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. tea cup\nB. telephone\nC. papers\nD. picture", "choices": ["A. tea cup", "B. telephone", "C. papers", "D. picture"], "answer": "C. papers", "target": "pen", "split": "validation"}
{"id": "c8d099ecd8_6", "scene_id": "c8d099ecd8", "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 papers, about 7 cm away. It is further next to the telephone, about 10 cm away.", "target": "pen", "split": "validation"}
{"id": "c8d099ecd8_7", "scene_id": "c8d099ecd8", "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\": \"papers\", \"relation\": \"next_to\", \"distance_cm\": 6.7}, {\"object\": \"telephone\", \"relation\": \"next_to\", \"distance_cm\": 9.8}]}", "target": "pen", "split": "validation"}
{"id": "ca0c580422_0", "scene_id": "ca0c580422", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "ca0c580422_1", "scene_id": "ca0c580422", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. post it notes\nB. monitor\nC. office chair\nD. cup", "choices": ["A. post it notes", "B. monitor", "C. office chair", "D. cup"], "answer": "B. monitor", "target": "bottle", "split": "validation"}
{"id": "ca0c580422_2", "scene_id": "ca0c580422", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 17 cm away. It is further near the cup, about 20 cm away.", "target": "bottle", "split": "validation"}
{"id": "ca0c580422_3", "scene_id": "ca0c580422", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 16.6}, {\"object\": \"cup\", \"relation\": \"near\", \"distance_cm\": 19.5}]}", "target": "bottle", "split": "validation"}
{"id": "ca0c580422_4", "scene_id": "ca0c580422", "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": "No", "target": "pen", "split": "validation"}
{"id": "ca0c580422_5", "scene_id": "ca0c580422", "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. post it notes\nB. box\nC. bottle\nD. table", "choices": ["A. post it notes", "B. box", "C. bottle", "D. table"], "answer": "A. post it notes", "target": "cup", "split": "validation"}
{"id": "ca0c580422_6", "scene_id": "ca0c580422", "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 post it notes, about 15 cm away. It is further near the bottle, about 20 cm away.", "target": "cup", "split": "validation"}
{"id": "ca0c580422_7", "scene_id": "ca0c580422", "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\": \"post it notes\", \"relation\": \"next_to\", \"distance_cm\": 14.6}, {\"object\": \"bottle\", \"relation\": \"near\", \"distance_cm\": 19.5}]}", "target": "cup", "split": "validation"}
{"id": "ca0c580422_8", "scene_id": "ca0c580422", "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": "No", "target": "mouse pad", "split": "validation"}
{"id": "ca0c580422_9", "scene_id": "ca0c580422", "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 above the table, about 4 cm away. It is also to the side of and above the cable, about 5 cm away. It is further to the side of and below the window sill, about 9 cm away.", "target": "phone charger", "split": "validation"}
{"id": "ca0c580422_10", "scene_id": "ca0c580422", "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\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 4.1}, {\"object\": \"cable\", \"relation\": \"side_above\", \"distance_cm\": 4.6}]}", "target": "phone charger", "split": "validation"}
{"id": "ca0c580422_11", "scene_id": "ca0c580422", "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": "Yes", "target": "stapler", "split": "validation"}
{"id": "ca0c580422_12", "scene_id": "ca0c580422", "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 stapler in this scene?\nA. monitor\nB. post it notes\nC. table\nD. box", "choices": ["A. monitor", "B. post it notes", "C. table", "D. box"], "answer": "B. post it notes", "target": "stapler", "split": "validation"}
{"id": "ca0c580422_13", "scene_id": "ca0c580422", "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 stapler in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The stapler is near the post it notes, about 17 cm away. It is also to the side of and above the table, about 22 cm away. It is further to the side of and below the monitor, about 22 cm away.", "target": "stapler", "split": "validation"}
{"id": "ca0c580422_14", "scene_id": "ca0c580422", "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 stapler in structured JSON format.", "choices": null, "answer": "{\"target\": \"stapler\", \"support_surface\": \"table\", \"references\": [{\"object\": \"post it notes\", \"relation\": \"near\", \"distance_cm\": 16.9}, {\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 21.9}]}", "target": "stapler", "split": "validation"}
{"id": "ce12db9e81_0", "scene_id": "ce12db9e81", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "ce12db9e81_1", "scene_id": "ce12db9e81", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. heater\nB. light switches\nC. chair\nD. heat pipe", "choices": ["A. heater", "B. light switches", "C. chair", "D. heat pipe"], "answer": "B. light switches", "target": "adapter", "split": "validation"}
{"id": "ce12db9e81_2", "scene_id": "ce12db9e81", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 near the light switches, about 18 cm away. It is also to the side of and above the chair, about 50 cm away.", "target": "adapter", "split": "validation"}
{"id": "ce12db9e81_3", "scene_id": "ce12db9e81", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"chair\", \"references\": [{\"object\": \"light switches\", \"relation\": \"near\", \"distance_cm\": 18.3}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 49.7}]}", "target": "adapter", "split": "validation"}
{"id": "ce12db9e81_4", "scene_id": "ce12db9e81", "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 pad appear?", "choices": null, "answer": "No", "target": "mouse pad", "split": "validation"}
{"id": "ce12db9e81_5", "scene_id": "ce12db9e81", "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. power sockets\nB. heater\nC. heat pipe\nD. heater pipe", "choices": ["A. power sockets", "B. heater", "C. heat pipe", "D. heater pipe"], "answer": "B. heater", "target": "power adapter", "split": "validation"}
{"id": "ce12db9e81_6", "scene_id": "ce12db9e81", "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 heater, about 20 cm away. It is also to the side of and above the heat pipe, about 51 cm away. It is further to the side of and above the heater pipe, about 61 cm away.", "target": "power adapter", "split": "validation"}
{"id": "ce12db9e81_7", "scene_id": "ce12db9e81", "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\": \"power sockets\", \"references\": [{\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 19.7}, {\"object\": \"heat pipe\", \"relation\": \"side_above\", \"distance_cm\": 51.2}]}", "target": "power adapter", "split": "validation"}
{"id": "d6d9ddb03f_0", "scene_id": "d6d9ddb03f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "d6d9ddb03f_1", "scene_id": "d6d9ddb03f", "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 charger appear?", "choices": null, "answer": "No", "target": "charger", "split": "validation"}
{"id": "d6d9ddb03f_2", "scene_id": "d6d9ddb03f", "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 bottle in this scene?\nA. office chair\nB. box\nC. backpack\nD. packet", "choices": ["A. office chair", "B. box", "C. backpack", "D. packet"], "answer": "B. box", "target": "bottle", "split": "validation"}
{"id": "d6d9ddb03f_3", "scene_id": "d6d9ddb03f", "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 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 box, about 16 cm away. It is also to the side of and above the office chair, about 45 cm away. It is further near the packet, about 48 cm away.", "target": "bottle", "split": "validation"}
{"id": "d6d9ddb03f_4", "scene_id": "d6d9ddb03f", "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 bottle in structured JSON format.", "choices": null, "answer": "{\"target\": \"bottle\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"box\", \"relation\": \"side_above\", \"distance_cm\": 15.6}, {\"object\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 44.6}]}", "target": "bottle", "split": "validation"}
{"id": "d6d9ddb03f_5", "scene_id": "d6d9ddb03f", "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 smartphone appear?", "choices": null, "answer": "No", "target": "smartphone", "split": "validation"}
{"id": "d6d9ddb03f_6", "scene_id": "d6d9ddb03f", "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 cup in this scene?\nA. keyboard\nB. file\nC. monitor\nD. mouse", "choices": ["A. keyboard", "B. file", "C. monitor", "D. mouse"], "answer": "D. mouse", "target": "cup", "split": "validation"}
{"id": "d6d9ddb03f_7", "scene_id": "d6d9ddb03f", "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 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 mouse, about 13 cm away. It is further near the keyboard, about 15 cm away.", "target": "cup", "split": "validation"}
{"id": "d6d9ddb03f_8", "scene_id": "d6d9ddb03f", "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 cup in structured JSON format.", "choices": null, "answer": "{\"target\": \"cup\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 13.4}, {\"object\": \"keyboard\", \"relation\": \"near\", \"distance_cm\": 15.3}]}", "target": "cup", "split": "validation"}
{"id": "d6d9ddb03f_9", "scene_id": "d6d9ddb03f", "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 headset appear?", "choices": null, "answer": "Yes", "target": "headset", "split": "validation"}
{"id": "d6d9ddb03f_10", "scene_id": "d6d9ddb03f", "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 headset in this scene?\nA. power socket\nB. keyboard\nC. monitor\nD. table lamp", "choices": ["A. power socket", "B. keyboard", "C. monitor", "D. table lamp"], "answer": "A. power socket", "target": "headset", "split": "validation"}
{"id": "d6d9ddb03f_11", "scene_id": "d6d9ddb03f", "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 headset in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The headset is near the power socket, about 16 cm away. It is also to the side of and below the table lamp, about 17 cm away. It is further to the side of and below the monitor, about 20 cm away.", "target": "headset", "split": "validation"}
{"id": "d6d9ddb03f_12", "scene_id": "d6d9ddb03f", "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 headset in structured JSON format.", "choices": null, "answer": "{\"target\": \"headset\", \"support_surface\": \"table lamp\", \"references\": [{\"object\": \"power socket\", \"relation\": \"near\", \"distance_cm\": 15.5}, {\"object\": \"table lamp\", \"relation\": \"side_below\", \"distance_cm\": 16.9}]}", "target": "headset", "split": "validation"}
{"id": "d6d9ddb03f_13", "scene_id": "d6d9ddb03f", "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 pen holder appear?", "choices": null, "answer": "No", "target": "pen holder", "split": "validation"}
{"id": "d6d9ddb03f_14", "scene_id": "d6d9ddb03f", "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 keyboard in this scene?\nA. mouse\nB. cup\nC. office chair\nD. file", "choices": ["A. mouse", "B. cup", "C. office chair", "D. file"], "answer": "A. mouse", "target": "keyboard", "split": "validation"}
{"id": "d6d9ddb03f_15", "scene_id": "d6d9ddb03f", "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 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 near the cup, about 15 cm away.", "target": "keyboard", "split": "validation"}
{"id": "d6d9ddb03f_16", "scene_id": "d6d9ddb03f", "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 keyboard in structured JSON format.", "choices": null, "answer": "{\"target\": \"keyboard\", \"support_surface\": \"table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 5.7}, {\"object\": \"cup\", \"relation\": \"near\", \"distance_cm\": 15.3}]}", "target": "keyboard", "split": "validation"}
{"id": "d6d9ddb03f_17", "scene_id": "d6d9ddb03f", "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 mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "validation"}
{"id": "d6d9ddb03f_18", "scene_id": "d6d9ddb03f", "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 mouse in this scene?\nA. keyboard\nB. office chair\nC. file\nD. cup", "choices": ["A. keyboard", "B. office chair", "C. file", "D. cup"], "answer": "A. keyboard", "target": "mouse", "split": "validation"}
{"id": "d6d9ddb03f_19", "scene_id": "d6d9ddb03f", "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 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 next to the cup, about 13 cm away.", "target": "mouse", "split": "validation"}
{"id": "d6d9ddb03f_20", "scene_id": "d6d9ddb03f", "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 mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 5.7}, {\"object\": \"cup\", \"relation\": \"next_to\", \"distance_cm\": 13.4}]}", "target": "mouse", "split": "validation"}
{"id": "d6d9ddb03f_21", "scene_id": "d6d9ddb03f", "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 tv controller appear?", "choices": null, "answer": "No", "target": "tv controller", "split": "validation"}
{"id": "d6d9ddb03f_22", "scene_id": "d6d9ddb03f", "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 notebook in this scene?\nA. cup\nB. cabinet\nC. file\nD. box", "choices": ["A. cup", "B. cabinet", "C. file", "D. box"], "answer": "D. box", "target": "notebook", "split": "validation"}
{"id": "d6d9ddb03f_23", "scene_id": "d6d9ddb03f", "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 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 below the box, about 7 cm away. It is further to the side of and above the cabinet, about 24 cm away.", "target": "notebook", "split": "validation"}
{"id": "d6d9ddb03f_24", "scene_id": "d6d9ddb03f", "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 notebook in structured JSON format.", "choices": null, "answer": "{\"target\": \"notebook\", \"support_surface\": \"table\", \"references\": [{\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 7.0}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 24.3}]}", "target": "notebook", "split": "validation"}
{"id": "d6d9ddb03f_25", "scene_id": "d6d9ddb03f", "source_dataset": "scannetpp", "local_sample_id": "25", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "d6d9ddb03f_26", "scene_id": "d6d9ddb03f", "source_dataset": "scannetpp", "local_sample_id": "26", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. table\nC. box\nD. power socket", "choices": ["A. heater", "B. table", "C. box", "D. power socket"], "answer": "A. heater", "target": "power strip", "split": "validation"}
{"id": "d6d9ddb03f_27", "scene_id": "d6d9ddb03f", "source_dataset": "scannetpp", "local_sample_id": "27", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 heater, about 41 cm away. It is also to the side of and below the box, about 47 cm away. It is further to the side of and below the table, about 51 cm away.", "target": "power strip", "split": "validation"}
{"id": "d6d9ddb03f_28", "scene_id": "d6d9ddb03f", "source_dataset": "scannetpp", "local_sample_id": "28", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"heater\", \"relation\": \"side_below\", \"distance_cm\": 41.4}, {\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 46.7}]}", "target": "power strip", "split": "validation"}
{"id": "d6d9ddb03f_29", "scene_id": "d6d9ddb03f", "source_dataset": "scannetpp", "local_sample_id": "29", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "d6d9ddb03f_30", "scene_id": "d6d9ddb03f", "source_dataset": "scannetpp", "local_sample_id": "30", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. jacket\nB. umbrella\nC. suitcase\nD. storage cabinet", "choices": ["A. jacket", "B. umbrella", "C. suitcase", "D. storage cabinet"], "answer": "D. storage cabinet", "target": "remote control", "split": "validation"}
{"id": "d6d9ddb03f_31", "scene_id": "d6d9ddb03f", "source_dataset": "scannetpp", "local_sample_id": "31", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 cabinet. It is also to the side of and below the storage cabinet, about 5 cm away. It is further to the side of and below the jacket, about 15 cm away.", "target": "remote control", "split": "validation"}
{"id": "d6d9ddb03f_32", "scene_id": "d6d9ddb03f", "source_dataset": "scannetpp", "local_sample_id": "32", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"cabinet\", \"references\": [{\"object\": \"storage cabinet\", \"relation\": \"side_below\", \"distance_cm\": 5.1}, {\"object\": \"jacket\", \"relation\": \"side_below\", \"distance_cm\": 14.9}]}", "target": "remote control", "split": "validation"}
{"id": "d918af9c5f_0", "scene_id": "d918af9c5f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "d918af9c5f_1", "scene_id": "d918af9c5f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. keyboard\nC. wheel\nD. picture frame", "choices": ["A. monitor", "B. keyboard", "C. wheel", "D. picture frame"], "answer": "D. picture frame", "target": "headphones", "split": "validation"}
{"id": "d918af9c5f_2", "scene_id": "d918af9c5f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 picture frame, about 4 cm away. It is further to the side of and below the monitor, about 12 cm away.", "target": "headphones", "split": "validation"}
{"id": "d918af9c5f_3", "scene_id": "d918af9c5f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"picture frame\", \"relation\": \"next_to\", \"distance_cm\": 4.1}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 12.5}]}", "target": "headphones", "split": "validation"}
{"id": "d918af9c5f_4", "scene_id": "d918af9c5f", "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": "validation"}
{"id": "d918af9c5f_5", "scene_id": "d918af9c5f", "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 desk. 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 laptop, about 8 cm away.", "target": "keyboard", "split": "validation"}
{"id": "d918af9c5f_6", "scene_id": "d918af9c5f", "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\": \"desk\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 8.2}, {\"object\": \"laptop\", \"relation\": \"side_below\", \"distance_cm\": 8.3}]}", "target": "keyboard", "split": "validation"}
{"id": "d918af9c5f_7", "scene_id": "d918af9c5f", "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 mouse appear?", "choices": null, "answer": "Yes", "target": "mouse", "split": "validation"}
{"id": "d918af9c5f_8", "scene_id": "d918af9c5f", "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. chair\nC. laptop\nD. desk lamp", "choices": ["A. keyboard", "B. chair", "C. laptop", "D. desk lamp"], "answer": "C. laptop", "target": "mouse", "split": "validation"}
{"id": "d918af9c5f_9", "scene_id": "d918af9c5f", "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 desk. It is also to the side of and below the laptop, about 11 cm away. It is further next to the keyboard, about 14 cm away.", "target": "mouse", "split": "validation"}
{"id": "d918af9c5f_10", "scene_id": "d918af9c5f", "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\": \"side_below\", \"distance_cm\": 10.8}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 13.6}]}", "target": "mouse", "split": "validation"}
{"id": "d918af9c5f_11", "scene_id": "d918af9c5f", "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": "validation"}
{"id": "d918af9c5f_12", "scene_id": "d918af9c5f", "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. bathroom cabinet\nB. tooth paste\nC. power sockets\nD. shelf", "choices": ["A. bathroom cabinet", "B. tooth paste", "C. power sockets", "D. shelf"], "answer": "B. tooth paste", "target": "toothbrush", "split": "validation"}
{"id": "d918af9c5f_13", "scene_id": "d918af9c5f", "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 tooth paste. 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 power sockets, about 13 cm away.", "target": "toothbrush", "split": "validation"}
{"id": "d918af9c5f_14", "scene_id": "d918af9c5f", "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\": \"tooth paste\", \"relation\": \"next_to\", \"distance_cm\": 2.1}, {\"object\": \"shelf\", \"relation\": \"side_below\", \"distance_cm\": 11.7}]}", "target": "toothbrush", "split": "validation"}
{"id": "dbd97b5bca_0", "scene_id": "dbd97b5bca", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "dbd97b5bca_1", "scene_id": "dbd97b5bca", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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 rack\nB. toilet paper\nC. bottle\nD. shelf rail", "choices": ["A. storage rack", "B. toilet paper", "C. bottle", "D. shelf rail"], "answer": "D. shelf rail", "target": "backpack", "split": "validation"}
{"id": "dbd97b5bca_2", "scene_id": "dbd97b5bca", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 shelf rail, about 3 cm away. It is also to the side of and below the bottle, about 11 cm away. It is further next to the storage rack, about 11 cm away.", "target": "backpack", "split": "validation"}
{"id": "dbd97b5bca_3", "scene_id": "dbd97b5bca", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"under\", \"distance_cm\": 3.1}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 10.8}]}", "target": "backpack", "split": "validation"}
{"id": "dbd97b5bca_4", "scene_id": "dbd97b5bca", "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": "validation"}
{"id": "dbd97b5bca_5", "scene_id": "dbd97b5bca", "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. towel\nC. bedsheet\nD. window frame", "choices": ["A. heater", "B. towel", "C. bedsheet", "D. window frame"], "answer": "C. bedsheet", "target": "power strip", "split": "validation"}
{"id": "dbd97b5bca_6", "scene_id": "dbd97b5bca", "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 bedsheet, about 14 cm away. It is further to the side of and above the heater, about 25 cm away.", "target": "power strip", "split": "validation"}
{"id": "dbd97b5bca_7", "scene_id": "dbd97b5bca", "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\": \"bedsheet\", \"relation\": \"side_above\", \"distance_cm\": 13.8}, {\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 25.4}]}", "target": "power strip", "split": "validation"}
{"id": "dd636ce603_0", "scene_id": "dd636ce603", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "dd636ce603_1", "scene_id": "dd636ce603", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. intercom\nB. cleaning supply\nC. power socket\nD. doorframe", "choices": ["A. intercom", "B. cleaning supply", "C. power socket", "D. doorframe"], "answer": "B. cleaning supply", "target": "bottle", "split": "validation"}
{"id": "dd636ce603_2", "scene_id": "dd636ce603", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 cleaning supply, about 12 cm away. It is also to the side of and below the intercom, about 20 cm away. It is further to the side of and below the doorframe, about 25 cm away.", "target": "bottle", "split": "validation"}
{"id": "dd636ce603_3", "scene_id": "dd636ce603", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"cleaning supply\", \"relation\": \"side_above\", \"distance_cm\": 12.1}, {\"object\": \"intercom\", \"relation\": \"side_below\", \"distance_cm\": 19.5}]}", "target": "bottle", "split": "validation"}
{"id": "dd636ce603_4", "scene_id": "dd636ce603", "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": "validation"}
{"id": "dd636ce603_5", "scene_id": "dd636ce603", "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. jar\nC. milk carton\nD. fruits", "choices": ["A. monitor", "B. jar", "C. milk carton", "D. fruits"], "answer": "B. jar", "target": "mug", "split": "validation"}
{"id": "dd636ce603_6", "scene_id": "dd636ce603", "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 next to the jar. It is also to the side of and below the monitor, about 6 cm away. It is further next to the milk carton, about 7 cm away.", "target": "mug", "split": "validation"}
{"id": "dd636ce603_7", "scene_id": "dd636ce603", "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\": \"bookshelf\", \"references\": [{\"object\": \"jar\", \"relation\": \"next_to\", \"distance_cm\": 2.5}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 6.1}]}", "target": "mug", "split": "validation"}
{"id": "dd636ce603_8", "scene_id": "dd636ce603", "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": "validation"}
{"id": "dd636ce603_9", "scene_id": "dd636ce603", "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. mouse\nB. milk carton\nC. mug\nD. tissue", "choices": ["A. mouse", "B. milk carton", "C. mug", "D. tissue"], "answer": "A. mouse", "target": "notebook", "split": "validation"}
{"id": "dd636ce603_10", "scene_id": "dd636ce603", "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 next to the mouse, about 8 cm away. It is also to the side of and below the mug, about 11 cm away. It is further to the side of and below the tissue, about 16 cm away.", "target": "notebook", "split": "validation"}
{"id": "dd636ce603_11", "scene_id": "dd636ce603", "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\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 8.3}, {\"object\": \"mug\", \"relation\": \"side_below\", \"distance_cm\": 11.2}]}", "target": "notebook", "split": "validation"}
{"id": "dd636ce603_12", "scene_id": "dd636ce603", "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": "validation"}
{"id": "dd636ce603_13", "scene_id": "dd636ce603", "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. power adapter\nB. power strip\nC. table\nD. laptop", "choices": ["A. power adapter", "B. power strip", "C. table", "D. laptop"], "answer": "B. power strip", "target": "spray bottle", "split": "validation"}
{"id": "dd636ce603_14", "scene_id": "dd636ce603", "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 window, about 3 cm away. It is also to the side of and above the power strip, about 13 cm away. It is further to the side of and above the table, about 29 cm away.", "target": "spray bottle", "split": "validation"}
{"id": "dd636ce603_15", "scene_id": "dd636ce603", "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\": \"table\", \"references\": [{\"object\": \"window\", \"relation\": \"next_to\", \"distance_cm\": 3.2}, {\"object\": \"power strip\", \"relation\": \"side_above\", \"distance_cm\": 13.2}]}", "target": "spray bottle", "split": "validation"}
{"id": "de5881aa12_0", "scene_id": "de5881aa12", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "de5881aa12_1", "scene_id": "de5881aa12", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. speaker/projector\nC. pillow/cushion\nD. sofa", "choices": ["A. bedside table", "B. speaker/projector", "C. pillow/cushion", "D. sofa"], "answer": "D. sofa", "target": "backpack", "split": "validation"}
{"id": "de5881aa12_2", "scene_id": "de5881aa12", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 sofa, about 3 cm away. It is also to the side of and below the pillow/cushion, about 27 cm away. It is further to the side of and below the speaker/projector, about 32 cm away.", "target": "backpack", "split": "validation"}
{"id": "de5881aa12_3", "scene_id": "de5881aa12", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"sofa\", \"relation\": \"under\", \"distance_cm\": 2.8}, {\"object\": \"pillow/cushion\", \"relation\": \"side_below\", \"distance_cm\": 27.5}]}", "target": "backpack", "split": "validation"}
{"id": "de5881aa12_4", "scene_id": "de5881aa12", "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": "No", "target": "calculator", "split": "validation"}
{"id": "de5881aa12_5", "scene_id": "de5881aa12", "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 earphones in this scene?\nA. box\nB. chair\nC. stationery\nD. laptop", "choices": ["A. box", "B. chair", "C. stationery", "D. laptop"], "answer": "D. laptop", "target": "earphones", "split": "validation"}
{"id": "de5881aa12_6", "scene_id": "de5881aa12", "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 earphones in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The earphones is next to the laptop, about 5 cm away. It is also to the side of and above the chair, about 26 cm away. It is further to the side of and below the stationery, about 34 cm away.", "target": "earphones", "split": "validation"}
{"id": "de5881aa12_7", "scene_id": "de5881aa12", "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 earphones in structured JSON format.", "choices": null, "answer": "{\"target\": \"earphones\", \"support_surface\": \"table\", \"references\": [{\"object\": \"laptop\", \"relation\": \"next_to\", \"distance_cm\": 5.1}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 26.0}]}", "target": "earphones", "split": "validation"}
{"id": "dfac5b38df_0", "scene_id": "dfac5b38df", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "dfac5b38df_1", "scene_id": "dfac5b38df", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. office chair\nC. food pot\nD. coaster", "choices": ["A. cabinet", "B. office chair", "C. food pot", "D. coaster"], "answer": "D. coaster", "target": "bottle", "split": "validation"}
{"id": "dfac5b38df_2", "scene_id": "dfac5b38df", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 6 cm away. It is also to the side of and above the food pot, about 15 cm away. It is further to the side of and above the office chair, about 36 cm away.", "target": "bottle", "split": "validation"}
{"id": "dfac5b38df_3", "scene_id": "dfac5b38df", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"coaster\", \"relation\": \"side_above\", \"distance_cm\": 6.2}, {\"object\": \"food pot\", \"relation\": \"side_above\", \"distance_cm\": 15.4}]}", "target": "bottle", "split": "validation"}
{"id": "dfac5b38df_4", "scene_id": "dfac5b38df", "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": "validation"}
{"id": "dfac5b38df_5", "scene_id": "dfac5b38df", "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. power adapter\nB. monitor\nC. keyboard\nD. mug", "choices": ["A. power adapter", "B. monitor", "C. keyboard", "D. mug"], "answer": "B. monitor", "target": "headphones", "split": "validation"}
{"id": "dfac5b38df_6", "scene_id": "dfac5b38df", "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 7 cm away. It is also next to the mug, about 10 cm away. It is further next to the power adapter, about 12 cm away.", "target": "headphones", "split": "validation"}
{"id": "dfac5b38df_7", "scene_id": "dfac5b38df", "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\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 7.1}, {\"object\": \"mug\", \"relation\": \"next_to\", \"distance_cm\": 10.3}]}", "target": "headphones", "split": "validation"}
{"id": "dfac5b38df_8", "scene_id": "dfac5b38df", "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": "validation"}
{"id": "dfac5b38df_9", "scene_id": "dfac5b38df", "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. monitor\nD. headphones", "choices": ["A. mouse", "B. office chair", "C. monitor", "D. headphones"], "answer": "A. mouse", "target": "keyboard", "split": "validation"}
{"id": "dfac5b38df_10", "scene_id": "dfac5b38df", "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 next to the mouse, about 8 cm away. It is also to the side of and below the monitor, about 9 cm away. It is further next to the headphones, about 14 cm away.", "target": "keyboard", "split": "validation"}
{"id": "dfac5b38df_11", "scene_id": "dfac5b38df", "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\": \"cabinet\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 8.1}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 9.2}]}", "target": "keyboard", "split": "validation"}
{"id": "dfac5b38df_12", "scene_id": "dfac5b38df", "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": "validation"}
{"id": "dfac5b38df_13", "scene_id": "dfac5b38df", "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 next to the keyboard, about 8 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 office chair, about 31 cm away.", "target": "mouse", "split": "validation"}
{"id": "dfac5b38df_14", "scene_id": "dfac5b38df", "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 chair\", \"references\": [{\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 8.1}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 8.1}]}", "target": "mouse", "split": "validation"}
{"id": "dfac5b38df_15", "scene_id": "dfac5b38df", "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": "Yes", "target": "mug", "split": "validation"}
{"id": "dfac5b38df_16", "scene_id": "dfac5b38df", "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. office chair\nB. cabinet\nC. headphones\nD. power adapter", "choices": ["A. office chair", "B. cabinet", "C. headphones", "D. power adapter"], "answer": "C. headphones", "target": "mug", "split": "validation"}
{"id": "dfac5b38df_17", "scene_id": "dfac5b38df", "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 next to the headphones, about 10 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 21 cm away.", "target": "mug", "split": "validation"}
{"id": "dfac5b38df_18", "scene_id": "dfac5b38df", "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\": \"cabinet\", \"references\": [{\"object\": \"headphones\", \"relation\": \"next_to\", \"distance_cm\": 10.3}, {\"object\": \"cabinet\", \"relation\": \"side_above\", \"distance_cm\": 16.7}]}", "target": "mug", "split": "validation"}
{"id": "e0de253456_0", "scene_id": "e0de253456", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "e0de253456_1", "scene_id": "e0de253456", "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 under the tile wall, about 3 cm away. It is also next to the split, about 3 cm away. It is further under the cream tube, about 3 cm away.", "target": "soap", "split": "validation"}
{"id": "e0de253456_2", "scene_id": "e0de253456", "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\": \"tile wall\", \"references\": [{\"object\": \"tile wall\", \"relation\": \"under\", \"distance_cm\": 2.5}, {\"object\": \"split\", \"relation\": \"next_to\", \"distance_cm\": 2.7}]}", "target": "soap", "split": "validation"}
{"id": "e3b3b0d0c7_0", "scene_id": "e3b3b0d0c7", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "e3b3b0d0c7_1", "scene_id": "e3b3b0d0c7", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. keyboard\nC. doorframe\nD. monitor", "choices": ["A. table", "B. keyboard", "C. doorframe", "D. monitor"], "answer": "C. doorframe", "target": "headphones", "split": "validation"}
{"id": "e3b3b0d0c7_2", "scene_id": "e3b3b0d0c7", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 doorframe, about 16 cm away. It is also near the door, about 22 cm away.", "target": "headphones", "split": "validation"}
{"id": "e3b3b0d0c7_3", "scene_id": "e3b3b0d0c7", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"door\", \"references\": [{\"object\": \"doorframe\", \"relation\": \"side_above\", \"distance_cm\": 16.3}, {\"object\": \"door\", \"relation\": \"near\", \"distance_cm\": 21.6}]}", "target": "headphones", "split": "validation"}
{"id": "e3ef8b690b_0", "scene_id": "e3ef8b690b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "e3ef8b690b_1", "scene_id": "e3ef8b690b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. bouquet\nB. chair\nC. plate\nD. tablecloth", "choices": ["A. bouquet", "B. chair", "C. plate", "D. tablecloth"], "answer": "A. bouquet", "target": "cup", "split": "validation"}
{"id": "e3ef8b690b_2", "scene_id": "e3ef8b690b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 bouquet, about 5 cm away. It is further to the side of and above the chair, about 31 cm away.", "target": "cup", "split": "validation"}
{"id": "e3ef8b690b_3", "scene_id": "e3ef8b690b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"bouquet\", \"relation\": \"under\", \"distance_cm\": 4.9}, {\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 31.0}]}", "target": "cup", "split": "validation"}
{"id": "e3ef8b690b_4", "scene_id": "e3ef8b690b", "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": "validation"}
{"id": "e3ef8b690b_5", "scene_id": "e3ef8b690b", "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 cushion\nB. bouquet\nC. blanket\nD. pillow", "choices": ["A. sofa cushion", "B. bouquet", "C. blanket", "D. pillow"], "answer": "D. pillow", "target": "tv remote", "split": "validation"}
{"id": "e3ef8b690b_6", "scene_id": "e3ef8b690b", "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 to the side of and below the pillow, about 6 cm away. It is also to the side of and above the blanket, about 15 cm away. It is further to the side of and below the sofa cushion, about 26 cm away.", "target": "tv remote", "split": "validation"}
{"id": "e3ef8b690b_7", "scene_id": "e3ef8b690b", "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\": \"sofa\", \"references\": [{\"object\": \"pillow\", \"relation\": \"side_below\", \"distance_cm\": 5.8}, {\"object\": \"blanket\", \"relation\": \"side_above\", \"distance_cm\": 15.3}]}", "target": "tv remote", "split": "validation"}
{"id": "e51b753e48_0", "scene_id": "e51b753e48", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "e51b753e48_1", "scene_id": "e51b753e48", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. plant\nC. lounge chair\nD. heater", "choices": ["A. book", "B. plant", "C. lounge chair", "D. heater"], "answer": "A. book", "target": "mug", "split": "validation"}
{"id": "e51b753e48_2", "scene_id": "e51b753e48", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 book. It is also next to the plant, about 5 cm away. It is further to the side of and above the lounge chair, about 9 cm away.", "target": "mug", "split": "validation"}
{"id": "e51b753e48_3", "scene_id": "e51b753e48", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 2.2}, {\"object\": \"plant\", \"relation\": \"next_to\", \"distance_cm\": 5.0}]}", "target": "mug", "split": "validation"}
{"id": "e51b753e48_4", "scene_id": "e51b753e48", "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": "validation"}
{"id": "e51b753e48_5", "scene_id": "e51b753e48", "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. bed\nB. poster\nC. windowsill\nD. heater", "choices": ["A. bed", "B. poster", "C. windowsill", "D. heater"], "answer": "C. windowsill", "target": "tissue", "split": "validation"}
{"id": "e51b753e48_6", "scene_id": "e51b753e48", "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 below the windowsill, about 10 cm away. It is also to the side of and below the poster, about 15 cm away. It is further to the side of and above the bed, about 17 cm away.", "target": "tissue", "split": "validation"}
{"id": "e51b753e48_7", "scene_id": "e51b753e48", "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\": \"bedside table\", \"references\": [{\"object\": \"windowsill\", \"relation\": \"side_below\", \"distance_cm\": 10.0}, {\"object\": \"poster\", \"relation\": \"side_below\", \"distance_cm\": 15.0}]}", "target": "tissue", "split": "validation"}
{"id": "e6372539da_0", "scene_id": "e6372539da", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "e6372539da_1", "scene_id": "e6372539da", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. washing machine\nD. light switches", "choices": ["A. cabinet", "B. bag", "C. washing machine", "D. light switches"], "answer": "C. washing machine", "target": "bottle", "split": "validation"}
{"id": "e6372539da_2", "scene_id": "e6372539da", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 washing machine, about 35 cm away. It is also to the side of and below the light switches, about 73 cm away.", "target": "bottle", "split": "validation"}
{"id": "e6372539da_3", "scene_id": "e6372539da", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"washing machine\", \"relation\": \"side_below\", \"distance_cm\": 34.9}, {\"object\": \"light switches\", \"relation\": \"side_below\", \"distance_cm\": 72.7}]}", "target": "bottle", "split": "validation"}
{"id": "e6372539da_4", "scene_id": "e6372539da", "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": "validation"}
{"id": "e6372539da_5", "scene_id": "e6372539da", "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. towel\nB. bag\nC. cloth\nD. drying rack", "choices": ["A. towel", "B. bag", "C. cloth", "D. drying rack"], "answer": "D. drying rack", "target": "power strip", "split": "validation"}
{"id": "e6372539da_6", "scene_id": "e6372539da", "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 next to the drying rack, about 12 cm away. It is also to the side of and above the bag, about 33 cm away. It is further to the side of and below the towel, about 51 cm away.", "target": "power strip", "split": "validation"}
{"id": "e6372539da_7", "scene_id": "e6372539da", "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\": \"drying rack\", \"references\": [{\"object\": \"drying rack\", \"relation\": \"next_to\", \"distance_cm\": 11.7}, {\"object\": \"bag\", \"relation\": \"side_above\", \"distance_cm\": 33.2}]}", "target": "power strip", "split": "validation"}
{"id": "e667e09fe6_0", "scene_id": "e667e09fe6", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "e667e09fe6_1", "scene_id": "e667e09fe6", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. bedside cabinet\nB. mug\nC. wardrobe\nD. shampoo bottle", "choices": ["A. bedside cabinet", "B. mug", "C. wardrobe", "D. shampoo bottle"], "answer": "A. bedside cabinet", "target": "headphones", "split": "validation"}
{"id": "e667e09fe6_2", "scene_id": "e667e09fe6", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 bedside cabinet, about 5 cm away. It is also to the side of and below the wardrobe, about 10 cm away. It is further to the side of and above the mug, about 12 cm away.", "target": "headphones", "split": "validation"}
{"id": "e667e09fe6_3", "scene_id": "e667e09fe6", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"bedside cabinet\", \"references\": [{\"object\": \"bedside cabinet\", \"relation\": \"side_above\", \"distance_cm\": 5.5}, {\"object\": \"wardrobe\", \"relation\": \"side_below\", \"distance_cm\": 10.0}]}", "target": "headphones", "split": "validation"}
{"id": "e667e09fe6_4", "scene_id": "e667e09fe6", "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": "validation"}
{"id": "e667e09fe6_5", "scene_id": "e667e09fe6", "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. usb hub\nB. laptop\nC. mousepad\nD. mouse", "choices": ["A. usb hub", "B. laptop", "C. mousepad", "D. mouse"], "answer": "C. mousepad", "target": "keyboard", "split": "validation"}
{"id": "e667e09fe6_6", "scene_id": "e667e09fe6", "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 mousepad, about 4 cm away. It is also next to the usb hub, about 8 cm away. It is further next to the mouse, about 9 cm away.", "target": "keyboard", "split": "validation"}
{"id": "e667e09fe6_7", "scene_id": "e667e09fe6", "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\": \"mousepad\", \"relation\": \"next_to\", \"distance_cm\": 3.6}, {\"object\": \"usb hub\", \"relation\": \"next_to\", \"distance_cm\": 7.8}]}", "target": "keyboard", "split": "validation"}
{"id": "e667e09fe6_8", "scene_id": "e667e09fe6", "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": "validation"}
{"id": "e667e09fe6_9", "scene_id": "e667e09fe6", "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. plastic bin\nB. trash can\nC. coffe table\nD. electric kettle", "choices": ["A. plastic bin", "B. trash can", "C. coffe table", "D. electric kettle"], "answer": "C. coffe table", "target": "laptop charger", "split": "validation"}
{"id": "e667e09fe6_10", "scene_id": "e667e09fe6", "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 coffe table, about 6 cm away. It is also to the side of and below the trash can, about 23 cm away. It is further to the side of and below the electric kettle, about 36 cm away.", "target": "laptop charger", "split": "validation"}
{"id": "e667e09fe6_11", "scene_id": "e667e09fe6", "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\": \"coffe table\", \"references\": [{\"object\": \"coffe table\", \"relation\": \"side_below\", \"distance_cm\": 5.8}, {\"object\": \"trash can\", \"relation\": \"side_below\", \"distance_cm\": 22.9}]}", "target": "laptop charger", "split": "validation"}
{"id": "e667e09fe6_12", "scene_id": "e667e09fe6", "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": "validation"}
{"id": "e667e09fe6_13", "scene_id": "e667e09fe6", "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. office chair\nC. keyboard\nD. glass", "choices": ["A. laptop", "B. office chair", "C. keyboard", "D. glass"], "answer": "B. office chair", "target": "mouse", "split": "validation"}
{"id": "e667e09fe6_14", "scene_id": "e667e09fe6", "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 7 cm away. It is further next to the keyboard, about 9 cm away.", "target": "mouse", "split": "validation"}
{"id": "e667e09fe6_15", "scene_id": "e667e09fe6", "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\": 7.0}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 8.6}]}", "target": "mouse", "split": "validation"}
{"id": "e667e09fe6_16", "scene_id": "e667e09fe6", "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 mousepad appear?", "choices": null, "answer": "Yes", "target": "mousepad", "split": "validation"}
{"id": "e667e09fe6_17", "scene_id": "e667e09fe6", "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 mousepad in this scene?\nA. laptop\nB. keyboard\nC. office chair\nD. glass", "choices": ["A. laptop", "B. keyboard", "C. office chair", "D. glass"], "answer": "D. glass", "target": "mousepad", "split": "validation"}
{"id": "e667e09fe6_18", "scene_id": "e667e09fe6", "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 mousepad in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The mousepad is under the glass. It is also next to the keyboard, about 4 cm away. It is further to the side of and below the laptop, about 6 cm away.", "target": "mousepad", "split": "validation"}
{"id": "e667e09fe6_19", "scene_id": "e667e09fe6", "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 mousepad in structured JSON format.", "choices": null, "answer": "{\"target\": \"mousepad\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"glass\", \"relation\": \"under\", \"distance_cm\": 2.3}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 3.6}]}", "target": "mousepad", "split": "validation"}
{"id": "e667e09fe6_20", "scene_id": "e667e09fe6", "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 toothbrush appear?", "choices": null, "answer": "No", "target": "toothbrush", "split": "validation"}
{"id": "e667e09fe6_21", "scene_id": "e667e09fe6", "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. poster\nB. bedside cabinet\nC. headphones\nD. shampoo bottle", "choices": ["A. poster", "B. bedside cabinet", "C. headphones", "D. shampoo bottle"], "answer": "C. headphones", "target": "mug", "split": "validation"}
{"id": "e667e09fe6_22", "scene_id": "e667e09fe6", "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 to the side of and below the headphones, about 12 cm away. It is also below the poster, about 25 cm away. It is further to the side of and below the shampoo bottle, about 26 cm away.", "target": "mug", "split": "validation"}
{"id": "e667e09fe6_23", "scene_id": "e667e09fe6", "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\": \"bedside cabinet\", \"references\": [{\"object\": \"headphones\", \"relation\": \"side_below\", \"distance_cm\": 11.5}, {\"object\": \"poster\", \"relation\": \"below\", \"distance_cm\": 25.4}]}", "target": "mug", "split": "validation"}
{"id": "e667e09fe6_24", "scene_id": "e667e09fe6", "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": "validation"}
{"id": "e667e09fe6_25", "scene_id": "e667e09fe6", "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 phone charger in this scene?\nA. office chair\nB. desk\nC. reusable shopping bag\nD. chair", "choices": ["A. office chair", "B. desk", "C. reusable shopping bag", "D. chair"], "answer": "D. chair", "target": "phone charger", "split": "validation"}
{"id": "e667e09fe6_26", "scene_id": "e667e09fe6", "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 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 chair, about 17 cm away. It is also near the rug, about 20 cm away. It is further to the side of and below the desk, about 29 cm away.", "target": "phone charger", "split": "validation"}
{"id": "e667e09fe6_27", "scene_id": "e667e09fe6", "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 phone charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"phone charger\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 16.5}, {\"object\": \"rug\", \"relation\": \"near\", \"distance_cm\": 19.5}]}", "target": "phone charger", "split": "validation"}
{"id": "e667e09fe6_28", "scene_id": "e667e09fe6", "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 tote bag appear?", "choices": null, "answer": "No", "target": "tote bag", "split": "validation"}
{"id": "e667e09fe6_29", "scene_id": "e667e09fe6", "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 usb hub in this scene?\nA. laptop\nB. monitor\nC. controller\nD. keyboard", "choices": ["A. laptop", "B. monitor", "C. controller", "D. keyboard"], "answer": "C. controller", "target": "usb hub", "split": "validation"}
{"id": "e667e09fe6_30", "scene_id": "e667e09fe6", "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 usb hub in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The usb hub is next to the controller. 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": "usb hub", "split": "validation"}
{"id": "e667e09fe6_31", "scene_id": "e667e09fe6", "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 usb hub in structured JSON format.", "choices": null, "answer": "{\"target\": \"usb hub\", \"support_surface\": \"desk\", \"references\": [{\"object\": \"controller\", \"relation\": \"next_to\", \"distance_cm\": 2.5}, {\"object\": \"keyboard\", \"relation\": \"next_to\", \"distance_cm\": 7.8}]}", "target": "usb hub", "split": "validation"}
{"id": "e7d7e7799d_0", "scene_id": "e7d7e7799d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "e7d7e7799d_1", "scene_id": "e7d7e7799d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. table\nB. blanket\nC. table lamp\nD. pillow", "choices": ["A. table", "B. blanket", "C. table lamp", "D. pillow"], "answer": "A. table", "target": "tv remote", "split": "validation"}
{"id": "e7d7e7799d_2", "scene_id": "e7d7e7799d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 table, about 12 cm away. It is also to the side of and above the blanket, about 17 cm away. It is further to the side of and below the pillow, about 17 cm away.", "target": "tv remote", "split": "validation"}
{"id": "e7d7e7799d_3", "scene_id": "e7d7e7799d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"bedside cabinet\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 12.2}, {\"object\": \"blanket\", \"relation\": \"side_above\", \"distance_cm\": 16.7}]}", "target": "tv remote", "split": "validation"}
{"id": "e8ea9b4da8_0", "scene_id": "e8ea9b4da8", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "e8ea9b4da8_1", "scene_id": "e8ea9b4da8", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. light switch\nB. bowl\nC. decoration\nD. book", "choices": ["A. light switch", "B. bowl", "C. decoration", "D. book"], "answer": "D. book", "target": "wallet", "split": "validation"}
{"id": "e8ea9b4da8_2", "scene_id": "e8ea9b4da8", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 book, about 3 cm away. It is also near the decoration, about 17 cm away. It is further to the side of and above the light switch, about 31 cm away.", "target": "wallet", "split": "validation"}
{"id": "e8ea9b4da8_3", "scene_id": "e8ea9b4da8", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"storage cabinet\", \"references\": [{\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 2.6}, {\"object\": \"decoration\", \"relation\": \"near\", \"distance_cm\": 17.4}]}", "target": "wallet", "split": "validation"}
{"id": "e909f8213d_0", "scene_id": "e909f8213d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "e909f8213d_1", "scene_id": "e909f8213d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. bag\nB. heater\nC. cutting board\nD. doorframe", "choices": ["A. bag", "B. heater", "C. cutting board", "D. doorframe"], "answer": "C. cutting board", "target": "bottle", "split": "validation"}
{"id": "e909f8213d_2", "scene_id": "e909f8213d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 cutting board, about 3 cm away. It is further to the side of and below the doorframe, about 14 cm away.", "target": "bottle", "split": "validation"}
{"id": "e909f8213d_3", "scene_id": "e909f8213d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"cutting board\", \"relation\": \"side_above\", \"distance_cm\": 3.0}, {\"object\": \"doorframe\", \"relation\": \"side_below\", \"distance_cm\": 13.7}]}", "target": "bottle", "split": "validation"}
{"id": "e909f8213d_4", "scene_id": "e909f8213d", "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": "validation"}
{"id": "e909f8213d_5", "scene_id": "e909f8213d", "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. sink\nB. sponge\nC. faucet\nD. toaster", "choices": ["A. sink", "B. sponge", "C. faucet", "D. toaster"], "answer": "B. sponge", "target": "cup", "split": "validation"}
{"id": "e909f8213d_6", "scene_id": "e909f8213d", "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 sponge, about 8 cm away. It is also next to the faucet, about 12 cm away. It is further to the side of and above the sink, about 14 cm away.", "target": "cup", "split": "validation"}
{"id": "e909f8213d_7", "scene_id": "e909f8213d", "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\": \"kitchen counter\", \"references\": [{\"object\": \"sponge\", \"relation\": \"side_above\", \"distance_cm\": 8.1}, {\"object\": \"faucet\", \"relation\": \"next_to\", \"distance_cm\": 12.0}]}", "target": "cup", "split": "validation"}
{"id": "e91722b5a3_0", "scene_id": "e91722b5a3", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "e91722b5a3_1", "scene_id": "e91722b5a3", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. envelope\nB. monitor\nC. paper\nD. keyboard", "choices": ["A. envelope", "B. monitor", "C. paper", "D. keyboard"], "answer": "B. monitor", "target": "stapler", "split": "validation"}
{"id": "e91722b5a3_2", "scene_id": "e91722b5a3", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 9 cm away. It is further next to the paper, about 13 cm away.", "target": "stapler", "split": "validation"}
{"id": "e91722b5a3_3", "scene_id": "e91722b5a3", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 9.0}, {\"object\": \"paper\", \"relation\": \"next_to\", \"distance_cm\": 13.4}]}", "target": "stapler", "split": "validation"}
{"id": "e93cd5230e_0", "scene_id": "e93cd5230e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "e93cd5230e_1", "scene_id": "e93cd5230e", "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 laptop charger appear?", "choices": null, "answer": "Yes", "target": "laptop charger", "split": "validation"}
{"id": "e93cd5230e_2", "scene_id": "e93cd5230e", "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 laptop charger in this scene?\nA. backpack\nB. heater\nC. power strip\nD. bag", "choices": ["A. backpack", "B. heater", "C. power strip", "D. bag"], "answer": "C. power strip", "target": "laptop charger", "split": "validation"}
{"id": "e93cd5230e_3", "scene_id": "e93cd5230e", "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 laptop charger in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The laptop charger is next to the power strip, about 5 cm away. It is also near the bag, about 46 cm away. It is further to the side of and below the heater, about 50 cm away.", "target": "laptop charger", "split": "validation"}
{"id": "e93cd5230e_4", "scene_id": "e93cd5230e", "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 laptop charger in structured JSON format.", "choices": null, "answer": "{\"target\": \"laptop charger\", \"support_surface\": \"bag\", \"references\": [{\"object\": \"power strip\", \"relation\": \"next_to\", \"distance_cm\": 5.2}, {\"object\": \"bag\", \"relation\": \"near\", \"distance_cm\": 46.1}]}", "target": "laptop charger", "split": "validation"}
{"id": "e93cd5230e_5", "scene_id": "e93cd5230e", "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 plug appear?", "choices": null, "answer": "No", "target": "power plug", "split": "validation"}
{"id": "e93cd5230e_6", "scene_id": "e93cd5230e", "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 mouse in this scene?\nA. table\nB. heater\nC. windowsill\nD. laptop", "choices": ["A. table", "B. heater", "C. windowsill", "D. laptop"], "answer": "A. table", "target": "mouse", "split": "validation"}
{"id": "e93cd5230e_7", "scene_id": "e93cd5230e", "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 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 near the laptop, about 15 cm away. It is further to the side of and above the heater, about 31 cm away.", "target": "mouse", "split": "validation"}
{"id": "e93cd5230e_8", "scene_id": "e93cd5230e", "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 mouse in structured JSON format.", "choices": null, "answer": "{\"target\": \"mouse\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_above\", \"distance_cm\": 4.2}, {\"object\": \"laptop\", \"relation\": \"near\", \"distance_cm\": 15.0}]}", "target": "mouse", "split": "validation"}
{"id": "e93cd5230e_9", "scene_id": "e93cd5230e", "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 mug appear?", "choices": null, "answer": "Yes", "target": "mug", "split": "validation"}
{"id": "e93cd5230e_10", "scene_id": "e93cd5230e", "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 mug in this scene?\nA. power sockets\nB. router\nC. coax outlet\nD. bottle", "choices": ["A. power sockets", "B. router", "C. coax outlet", "D. bottle"], "answer": "D. bottle", "target": "mug", "split": "validation"}
{"id": "e93cd5230e_11", "scene_id": "e93cd5230e", "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 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 window, about 12 cm away. It is further to the side of and below the bottle, about 29 cm away.", "target": "mug", "split": "validation"}
{"id": "e93cd5230e_12", "scene_id": "e93cd5230e", "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 mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"table\", \"references\": [{\"object\": \"window\", \"relation\": \"side_below\", \"distance_cm\": 11.7}, {\"object\": \"bottle\", \"relation\": \"side_below\", \"distance_cm\": 29.2}]}", "target": "mug", "split": "validation"}
{"id": "e93cd5230e_13", "scene_id": "e93cd5230e", "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 toothbrush appear?", "choices": null, "answer": "No", "target": "toothbrush", "split": "validation"}
{"id": "e93cd5230e_14", "scene_id": "e93cd5230e", "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. laptop charger\nB. backpack\nC. bag\nD. heater", "choices": ["A. laptop charger", "B. backpack", "C. bag", "D. heater"], "answer": "A. laptop charger", "target": "power strip", "split": "validation"}
{"id": "e93cd5230e_15", "scene_id": "e93cd5230e", "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 next to the laptop charger, about 5 cm away. It is also near the bag, about 44 cm away. It is further to the side of and below the backpack, about 47 cm away.", "target": "power strip", "split": "validation"}
{"id": "e93cd5230e_16", "scene_id": "e93cd5230e", "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\": \"bag\", \"references\": [{\"object\": \"laptop charger\", \"relation\": \"next_to\", \"distance_cm\": 5.2}, {\"object\": \"bag\", \"relation\": \"near\", \"distance_cm\": 44.0}]}", "target": "power strip", "split": "validation"}
{"id": "ed2216380b_0", "scene_id": "ed2216380b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "ed2216380b_1", "scene_id": "ed2216380b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. chair\nD. envelope", "choices": ["A. box", "B. bottle", "C. chair", "D. envelope"], "answer": "C. chair", "target": "backpack", "split": "validation"}
{"id": "ed2216380b_2", "scene_id": "ed2216380b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 envelope, about 51 cm away.", "target": "backpack", "split": "validation"}
{"id": "ed2216380b_3", "scene_id": "ed2216380b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"envelope\", \"relation\": \"side_below\", \"distance_cm\": 51.5}]}", "target": "backpack", "split": "validation"}
{"id": "ed2216380b_4", "scene_id": "ed2216380b", "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": "validation"}
{"id": "ed2216380b_5", "scene_id": "ed2216380b", "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. box\nB. whiteboard eraser\nC. backpack\nD. bottle", "choices": ["A. box", "B. whiteboard eraser", "C. backpack", "D. bottle"], "answer": "B. whiteboard eraser", "target": "pen holder", "split": "validation"}
{"id": "ed2216380b_6", "scene_id": "ed2216380b", "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 whiteboard eraser.", "target": "pen holder", "split": "validation"}
{"id": "ed2216380b_7", "scene_id": "ed2216380b", "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\": \"whiteboard eraser\", \"references\": [{\"object\": \"whiteboard eraser\", \"relation\": \"next_to\", \"distance_cm\": 2.4}]}", "target": "pen holder", "split": "validation"}
{"id": "eea4ad9c04_0", "scene_id": "eea4ad9c04", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "eea4ad9c04_1", "scene_id": "eea4ad9c04", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. cabinet\nC. lantern\nD. vase", "choices": ["A. chair", "B. cabinet", "C. lantern", "D. vase"], "answer": "B. cabinet", "target": "extension cord", "split": "validation"}
{"id": "eea4ad9c04_2", "scene_id": "eea4ad9c04", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 cabinet, about 22 cm away. It is also near the vase, about 32 cm away. It is further to the side of and below the chair, about 42 cm away.", "target": "extension cord", "split": "validation"}
{"id": "eea4ad9c04_3", "scene_id": "eea4ad9c04", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 22.1}, {\"object\": \"vase\", \"relation\": \"near\", \"distance_cm\": 32.1}]}", "target": "extension cord", "split": "validation"}
{"id": "eea4ad9c04_4", "scene_id": "eea4ad9c04", "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": "validation"}
{"id": "eea4ad9c04_5", "scene_id": "eea4ad9c04", "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. books\nB. phone charger\nC. vase\nD. cabinet", "choices": ["A. books", "B. phone charger", "C. vase", "D. cabinet"], "answer": "B. phone charger", "target": "laptop charger", "split": "validation"}
{"id": "eea4ad9c04_6", "scene_id": "eea4ad9c04", "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 cabinet, about 11 cm away. It is also to the side of and below the vase, about 21 cm away. It is further to the side of and below the books, about 56 cm away.", "target": "laptop charger", "split": "validation"}
{"id": "eea4ad9c04_7", "scene_id": "eea4ad9c04", "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\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 11.1}, {\"object\": \"vase\", \"relation\": \"side_below\", \"distance_cm\": 21.4}]}", "target": "laptop charger", "split": "validation"}
{"id": "eea4ad9c04_8", "scene_id": "eea4ad9c04", "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 appear?", "choices": null, "answer": "No", "target": "phone", "split": "validation"}
{"id": "eea4ad9c04_9", "scene_id": "eea4ad9c04", "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 next to the trolley, about 6 cm away. It is also to the side of and below the digital piano cover, about 7 cm away. It is further near the bedside table, about 24 cm away.", "target": "phone charger", "split": "validation"}
{"id": "eea4ad9c04_10", "scene_id": "eea4ad9c04", "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\": \"bedside table\", \"references\": [{\"object\": \"trolley\", \"relation\": \"next_to\", \"distance_cm\": 6.1}, {\"object\": \"digital piano cover\", \"relation\": \"side_below\", \"distance_cm\": 7.1}]}", "target": "phone charger", "split": "validation"}
{"id": "eee7eef8e5_0", "scene_id": "eee7eef8e5", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "eee7eef8e5_1", "scene_id": "eee7eef8e5", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. chair\nB. table\nC. monitor\nD. split", "choices": ["A. chair", "B. table", "C. monitor", "D. split"], "answer": "A. chair", "target": "remote", "split": "validation"}
{"id": "eee7eef8e5_2", "scene_id": "eee7eef8e5", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 chair, about 19 cm away.", "target": "remote", "split": "validation"}
{"id": "eee7eef8e5_3", "scene_id": "eee7eef8e5", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"chair\", \"references\": [{\"object\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 19.2}]}", "target": "remote", "split": "validation"}
{"id": "ef18cf0708_0", "scene_id": "ef18cf0708", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "ef18cf0708_1", "scene_id": "ef18cf0708", "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 cable appear?", "choices": null, "answer": "Yes", "target": "cable", "split": "validation"}
{"id": "ef18cf0708_2", "scene_id": "ef18cf0708", "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 cable in this scene?\nA. electrical plug\nB. monitor\nC. power socket\nD. mouse", "choices": ["A. electrical plug", "B. monitor", "C. power socket", "D. mouse"], "answer": "B. monitor", "target": "cable", "split": "validation"}
{"id": "ef18cf0708_3", "scene_id": "ef18cf0708", "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 cable in the scene, including its distance and relation to nearby objects.", "choices": null, "answer": "The cable is on the table. It is also under the monitor, about 3 cm away. It is further to the side of and below the electrical plug, about 19 cm away.", "target": "cable", "split": "validation"}
{"id": "ef18cf0708_4", "scene_id": "ef18cf0708", "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 cable in structured JSON format.", "choices": null, "answer": "{\"target\": \"cable\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"under\", \"distance_cm\": 2.9}, {\"object\": \"electrical plug\", \"relation\": \"side_below\", \"distance_cm\": 19.2}]}", "target": "cable", "split": "validation"}
{"id": "ef18cf0708_5", "scene_id": "ef18cf0708", "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 comb appear?", "choices": null, "answer": "No", "target": "comb", "split": "validation"}
{"id": "ef18cf0708_6", "scene_id": "ef18cf0708", "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. power socket\nB. windowsill\nC. window sill\nD. heater", "choices": ["A. power socket", "B. windowsill", "C. window sill", "D. heater"], "answer": "D. heater", "target": "keyboard", "split": "validation"}
{"id": "ef18cf0708_7", "scene_id": "ef18cf0708", "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 to the side of and above the heater, about 28 cm away. It is also to the side of and below the power socket, about 36 cm away. It is further to the side of and below the window sill, about 42 cm away.", "target": "keyboard", "split": "validation"}
{"id": "ef18cf0708_8", "scene_id": "ef18cf0708", "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\": \"shelf\", \"references\": [{\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 27.8}, {\"object\": \"power socket\", \"relation\": \"side_below\", \"distance_cm\": 36.5}]}", "target": "keyboard", "split": "validation"}
{"id": "ef18cf0708_9", "scene_id": "ef18cf0708", "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 mouse pad appear?", "choices": null, "answer": "No", "target": "mouse pad", "split": "validation"}
{"id": "ef18cf0708_10", "scene_id": "ef18cf0708", "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 mug in this scene?\nA. windowsill\nB. split\nC. telephone\nD. power socket", "choices": ["A. windowsill", "B. split", "C. telephone", "D. power socket"], "answer": "A. windowsill", "target": "mug", "split": "validation"}
{"id": "ef18cf0708_11", "scene_id": "ef18cf0708", "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 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 windowsill, about 3 cm away. It is also next to the telephone, about 6 cm away. It is further to the side of and above the power socket, about 21 cm away.", "target": "mug", "split": "validation"}
{"id": "ef18cf0708_12", "scene_id": "ef18cf0708", "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 mug in structured JSON format.", "choices": null, "answer": "{\"target\": \"mug\", \"support_surface\": \"shelf\", \"references\": [{\"object\": \"windowsill\", \"relation\": \"side_above\", \"distance_cm\": 3.0}, {\"object\": \"telephone\", \"relation\": \"next_to\", \"distance_cm\": 6.4}]}", "target": "mug", "split": "validation"}
{"id": "f19ca0a52e_0", "scene_id": "f19ca0a52e", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "f19ca0a52e_1", "scene_id": "f19ca0a52e", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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 outlet\nB. cutting board\nC. kettle\nD. dishwasher", "choices": ["A. power outlet", "B. cutting board", "C. kettle", "D. dishwasher"], "answer": "C. kettle", "target": "cup", "split": "validation"}
{"id": "f19ca0a52e_2", "scene_id": "f19ca0a52e", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 to the side of and below the kettle, about 6 cm away. It is further to the side of and above the dishwasher, about 32 cm away.", "target": "cup", "split": "validation"}
{"id": "f19ca0a52e_3", "scene_id": "f19ca0a52e", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"kettle\", \"relation\": \"side_below\", \"distance_cm\": 6.2}, {\"object\": \"dishwasher\", \"relation\": \"side_above\", \"distance_cm\": 32.2}]}", "target": "cup", "split": "validation"}
{"id": "f3d64c30f8_0", "scene_id": "f3d64c30f8", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "f3d64c30f8_1", "scene_id": "f3d64c30f8", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. jug\nB. box\nC. cabinet\nD. whiteboard", "choices": ["A. jug", "B. box", "C. cabinet", "D. whiteboard"], "answer": "B. box", "target": "cup", "split": "validation"}
{"id": "f3d64c30f8_2", "scene_id": "f3d64c30f8", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 box, about 7 cm away. It is also to the side of and above the jug, about 24 cm away. It is further to the side of and below the whiteboard, about 25 cm away.", "target": "cup", "split": "validation"}
{"id": "f3d64c30f8_3", "scene_id": "f3d64c30f8", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 7.2}, {\"object\": \"jug\", \"relation\": \"side_above\", \"distance_cm\": 24.2}]}", "target": "cup", "split": "validation"}
{"id": "f3d64c30f8_4", "scene_id": "f3d64c30f8", "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": "validation"}
{"id": "f3d64c30f8_5", "scene_id": "f3d64c30f8", "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. bookshelf\nB. power socket\nC. bag\nD. bottle", "choices": ["A. bookshelf", "B. power socket", "C. bag", "D. bottle"], "answer": "B. power socket", "target": "power strip", "split": "validation"}
{"id": "f3d64c30f8_6", "scene_id": "f3d64c30f8", "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 power socket, about 50 cm away.", "target": "power strip", "split": "validation"}
{"id": "f3d64c30f8_7", "scene_id": "f3d64c30f8", "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\": \"power socket\", \"references\": [{\"object\": \"power socket\", \"relation\": \"side_below\", \"distance_cm\": 50.3}]}", "target": "power strip", "split": "validation"}
{"id": "f3f016ba3f_0", "scene_id": "f3f016ba3f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "f3f016ba3f_1", "scene_id": "f3f016ba3f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. pen holder\nB. monitor\nC. toy\nD. office chair", "choices": ["A. pen holder", "B. monitor", "C. toy", "D. office chair"], "answer": "D. office chair", "target": "headphones", "split": "validation"}
{"id": "f3f016ba3f_2", "scene_id": "f3f016ba3f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 above the office chair, about 3 cm away. It is further to the side of and below the monitor, about 12 cm away.", "target": "headphones", "split": "validation"}
{"id": "f3f016ba3f_3", "scene_id": "f3f016ba3f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"office chair\", \"relation\": \"side_above\", \"distance_cm\": 3.0}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 11.7}]}", "target": "headphones", "split": "validation"}
{"id": "f3f016ba3f_4", "scene_id": "f3f016ba3f", "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": "validation"}
{"id": "f3f016ba3f_5", "scene_id": "f3f016ba3f", "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 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 monitor, about 10 cm away. It is further to the side of and above the toy, about 10 cm away.", "target": "pen holder", "split": "validation"}
{"id": "f3f016ba3f_6", "scene_id": "f3f016ba3f", "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 pen holder in structured JSON format.", "choices": null, "answer": "{\"target\": \"pen holder\", \"support_surface\": \"table\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 9.6}, {\"object\": \"toy\", \"relation\": \"side_above\", \"distance_cm\": 10.1}]}", "target": "pen holder", "split": "validation"}
{"id": "f3f016ba3f_7", "scene_id": "f3f016ba3f", "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 controller appear?", "choices": null, "answer": "No", "target": "tv controller", "split": "validation"}
{"id": "f3f016ba3f_8", "scene_id": "f3f016ba3f", "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 power strip in this scene?\nA. picture\nB. yoga mat\nC. toy\nD. table", "choices": ["A. picture", "B. yoga mat", "C. toy", "D. table"], "answer": "D. table", "target": "power strip", "split": "validation"}
{"id": "f3f016ba3f_9", "scene_id": "f3f016ba3f", "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 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 46 cm away. It is also near the yoga mat, about 48 cm away. It is further to the side of and below the toy, about 67 cm away.", "target": "power strip", "split": "validation"}
{"id": "f3f016ba3f_10", "scene_id": "f3f016ba3f", "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 power strip in structured JSON format.", "choices": null, "answer": "{\"target\": \"power strip\", \"support_surface\": \"table\", \"references\": [{\"object\": \"table\", \"relation\": \"side_below\", \"distance_cm\": 46.2}, {\"object\": \"yoga mat\", \"relation\": \"near\", \"distance_cm\": 47.5}]}", "target": "power strip", "split": "validation"}
{"id": "f6659a3107_0", "scene_id": "f6659a3107", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "f6659a3107_1", "scene_id": "f6659a3107", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. table\nB. keyboard\nC. split\nD. monitor", "choices": ["A. table", "B. keyboard", "C. split", "D. monitor"], "answer": "C. split", "target": "eraser", "split": "validation"}
{"id": "f6659a3107_2", "scene_id": "f6659a3107", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 split, about 26 cm away.", "target": "eraser", "split": "validation"}
{"id": "f6659a3107_3", "scene_id": "f6659a3107", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"split\", \"references\": [{\"object\": \"split\", \"relation\": \"side_above\", \"distance_cm\": 26.0}]}", "target": "eraser", "split": "validation"}
{"id": "f6a9b64a0d_0", "scene_id": "f6a9b64a0d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "f6a9b64a0d_1", "scene_id": "f6a9b64a0d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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 lamp\nB. table\nC. plug\nD. book", "choices": ["A. bedside lamp", "B. table", "C. plug", "D. book"], "answer": "B. table", "target": "power strip", "split": "validation"}
{"id": "f6a9b64a0d_2", "scene_id": "f6a9b64a0d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 book, about 30 cm away. It is further to the side of and below the bedside lamp, about 34 cm away.", "target": "power strip", "split": "validation"}
{"id": "f6a9b64a0d_3", "scene_id": "f6a9b64a0d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": 5.7}, {\"object\": \"book\", \"relation\": \"side_below\", \"distance_cm\": 30.2}]}", "target": "power strip", "split": "validation"}
{"id": "f8d5147d1d_0", "scene_id": "f8d5147d1d", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "f8d5147d1d_1", "scene_id": "f8d5147d1d", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. office table\nD. hole puncher", "choices": ["A. monitor", "B. stapler", "C. office table", "D. hole puncher"], "answer": "C. office table", "target": "power strip", "split": "validation"}
{"id": "f8d5147d1d_2", "scene_id": "f8d5147d1d", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 hole puncher, about 20 cm away. It is also to the side of and below the monitor, about 25 cm away. It is further to the side of and below the stapler, about 34 cm away.", "target": "power strip", "split": "validation"}
{"id": "f8d5147d1d_3", "scene_id": "f8d5147d1d", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"hole puncher\", \"relation\": \"side_below\", \"distance_cm\": 20.0}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 24.5}]}", "target": "power strip", "split": "validation"}
{"id": "f8d5147d1d_4", "scene_id": "f8d5147d1d", "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": "validation"}
{"id": "f8d5147d1d_5", "scene_id": "f8d5147d1d", "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. monitor\nB. mouse\nC. office table\nD. cabinet", "choices": ["A. monitor", "B. mouse", "C. office table", "D. cabinet"], "answer": "B. mouse", "target": "stapler", "split": "validation"}
{"id": "f8d5147d1d_6", "scene_id": "f8d5147d1d", "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 next to the mouse, about 12 cm away. It is also to the side of and below the monitor, about 22 cm away. It is further to the side of and above the cabinet, about 27 cm away.", "target": "stapler", "split": "validation"}
{"id": "f8d5147d1d_7", "scene_id": "f8d5147d1d", "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\": \"office table\", \"references\": [{\"object\": \"mouse\", \"relation\": \"next_to\", \"distance_cm\": 12.1}, {\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 22.1}]}", "target": "stapler", "split": "validation"}
{"id": "f8f12e4e6b_0", "scene_id": "f8f12e4e6b", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "f8f12e4e6b_1", "scene_id": "f8f12e4e6b", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. box\nC. split\nD. shirt", "choices": ["A. office chair", "B. box", "C. split", "D. shirt"], "answer": "B. box", "target": "keyboard", "split": "validation"}
{"id": "f8f12e4e6b_2", "scene_id": "f8f12e4e6b", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 box, about 3 cm away. It is further next to the split, about 14 cm away.", "target": "keyboard", "split": "validation"}
{"id": "f8f12e4e6b_3", "scene_id": "f8f12e4e6b", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"box\", \"relation\": \"next_to\", \"distance_cm\": 2.7}, {\"object\": \"split\", \"relation\": \"next_to\", \"distance_cm\": 13.8}]}", "target": "keyboard", "split": "validation"}
{"id": "f8f12e4e6b_4", "scene_id": "f8f12e4e6b", "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": "validation"}
{"id": "f8f12e4e6b_5", "scene_id": "f8f12e4e6b", "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. chair\nB. shirt\nC. keyboard\nD. box", "choices": ["A. chair", "B. shirt", "C. keyboard", "D. box"], "answer": "A. chair", "target": "tissue", "split": "validation"}
{"id": "f8f12e4e6b_6", "scene_id": "f8f12e4e6b", "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 on the table. It is also to the side of and above the chair, about 16 cm away. It is further near the shirt, about 19 cm away.", "target": "tissue", "split": "validation"}
{"id": "f8f12e4e6b_7", "scene_id": "f8f12e4e6b", "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\": \"chair\", \"relation\": \"side_above\", \"distance_cm\": 15.9}, {\"object\": \"shirt\", \"relation\": \"near\", \"distance_cm\": 18.6}]}", "target": "tissue", "split": "validation"}
{"id": "fb05e13ad1_0", "scene_id": "fb05e13ad1", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "fb05e13ad1_1", "scene_id": "fb05e13ad1", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. cabinet\nD. cushion", "choices": ["A. book", "B. chair", "C. cabinet", "D. cushion"], "answer": "D. cushion", "target": "backpack", "split": "validation"}
{"id": "fb05e13ad1_2", "scene_id": "fb05e13ad1", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 cushion, about 32 cm away.", "target": "backpack", "split": "validation"}
{"id": "fb05e13ad1_3", "scene_id": "fb05e13ad1", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"cushion\", \"relation\": \"side_below\", \"distance_cm\": 31.9}]}", "target": "backpack", "split": "validation"}
{"id": "fb05e13ad1_4", "scene_id": "fb05e13ad1", "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": "validation"}
{"id": "fb05e13ad1_5", "scene_id": "fb05e13ad1", "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. paper\nB. laptop\nC. book\nD. chair", "choices": ["A. paper", "B. laptop", "C. book", "D. chair"], "answer": "C. book", "target": "cup", "split": "validation"}
{"id": "fb05e13ad1_6", "scene_id": "fb05e13ad1", "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 book, about 11 cm away. It is further to the side of and above the laptop, about 14 cm away.", "target": "cup", "split": "validation"}
{"id": "fb05e13ad1_7", "scene_id": "fb05e13ad1", "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\": \"side_above\", \"distance_cm\": 10.8}, {\"object\": \"laptop\", \"relation\": \"side_above\", \"distance_cm\": 14.2}]}", "target": "cup", "split": "validation"}
{"id": "fb05e13ad1_8", "scene_id": "fb05e13ad1", "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": "No", "target": "mouse pad", "split": "validation"}
{"id": "fb05e13ad1_9", "scene_id": "fb05e13ad1", "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. power socket\nC. book\nD. telephone", "choices": ["A. monitor", "B. power socket", "C. book", "D. telephone"], "answer": "A. monitor", "target": "keyboard", "split": "validation"}
{"id": "fb05e13ad1_10", "scene_id": "fb05e13ad1", "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 6 cm away. It is also next to the book, about 9 cm away. It is further to the side of and below the telephone, about 9 cm away.", "target": "keyboard", "split": "validation"}
{"id": "fb05e13ad1_11", "scene_id": "fb05e13ad1", "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\": \"book\", \"references\": [{\"object\": \"monitor\", \"relation\": \"side_below\", \"distance_cm\": 5.7}, {\"object\": \"book\", \"relation\": \"next_to\", \"distance_cm\": 9.1}]}", "target": "keyboard", "split": "validation"}
{"id": "fe1733741f_0", "scene_id": "fe1733741f", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "Yes", "target": "phone", "split": "validation"}
{"id": "fe1733741f_1", "scene_id": "fe1733741f", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. plant\nB. books\nC. keyboard\nD. plant pot", "choices": ["A. plant", "B. books", "C. keyboard", "D. plant pot"], "answer": "B. books", "target": "phone", "split": "validation"}
{"id": "fe1733741f_2", "scene_id": "fe1733741f", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 below the window, about 6 cm away. It is also next to the books, about 7 cm away. It is further near the keyboard, about 27 cm away.", "target": "phone", "split": "validation"}
{"id": "fe1733741f_3", "scene_id": "fe1733741f", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"window\", \"references\": [{\"object\": \"window\", \"relation\": \"side_below\", \"distance_cm\": 5.6}, {\"object\": \"books\", \"relation\": \"next_to\", \"distance_cm\": 7.0}]}", "target": "phone", "split": "validation"}
{"id": "fe1733741f_4", "scene_id": "fe1733741f", "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": "validation"}
{"id": "fe1733741f_5", "scene_id": "fe1733741f", "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 tote bag in this scene?\nA. mug\nB. cabinet\nC. box\nD. books", "choices": ["A. mug", "B. cabinet", "C. box", "D. books"], "answer": "B. cabinet", "target": "tote bag", "split": "validation"}
{"id": "fe1733741f_6", "scene_id": "fe1733741f", "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 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 cabinet, about 5 cm away. It is also to the side of and below the box, about 42 cm away. It is further to the side of and below the mug, about 71 cm away.", "target": "tote bag", "split": "validation"}
{"id": "fe1733741f_7", "scene_id": "fe1733741f", "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 tote bag in structured JSON format.", "choices": null, "answer": "{\"target\": \"tote bag\", \"support_surface\": \"cabinet\", \"references\": [{\"object\": \"cabinet\", \"relation\": \"side_below\", \"distance_cm\": 5.3}, {\"object\": \"box\", \"relation\": \"side_below\", \"distance_cm\": 42.0}]}", "target": "tote bag", "split": "validation"}
{"id": "fe3fc057a1_0", "scene_id": "fe3fc057a1", "source_dataset": "scannetpp", "local_sample_id": "0", "task": "TPV", "question_type": "target 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": "validation"}
{"id": "fe3fc057a1_1", "scene_id": "fe3fc057a1", "source_dataset": "scannetpp", "local_sample_id": "1", "task": "NRI", "question_type": "nearest reference identification", "instruction": "Choose exactly one option from A B C D.\nOutput ONLY in the format: X. object_name\nDo not output 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. documents\nB. mug\nC. windowsill\nD. heater", "choices": ["A. documents", "B. mug", "C. windowsill", "D. heater"], "answer": "C. windowsill", "target": "mouse", "split": "validation"}
{"id": "fe3fc057a1_2", "scene_id": "fe3fc057a1", "source_dataset": "scannetpp", "local_sample_id": "2", "task": "FSD", "question_type": "fine-grained spatial description", "instruction": "Describe the location naturally in 1-2 sentences. Mention the support surface and 1-2 closest objects with exact distances in centimeters (e.g., 23 cm, 41 cm). 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 windowsill, about 6 cm away. It is further to the side of and above the heater, about 17 cm away.", "target": "mouse", "split": "validation"}
{"id": "fe3fc057a1_3", "scene_id": "fe3fc057a1", "source_dataset": "scannetpp", "local_sample_id": "3", "task": "SSP", "question_type": "structured spatial prediction", "instruction": "Output ONLY a valid JSON object. No extra text, no explanation, no markdown, no ```json.\n\nUse exactly these fields:\n- target: string (main object name)\n- support_surface: string (main surface or \"none\")\n- references: array of 0-2 objects, each with:\n  - object: string\n  - relation: one of on, under, above, below, side_above, side_below, next_to, near, attached_to\n  - distance_cm: float (e.g. 8.0 or 20.4)\n\nExample:\n{\"target\": \"backpack\", \"support_surface\": \"none\", \"references\": [\n  {\"object\": \"chair\", \"relation\": \"side_below\", \"distance_cm\": 17.0},\n  {\"object\": \"nightstand\", \"relation\": \"side_below\", \"distance_cm\": 19.0}\n]}\n\nRules:\n1. Max 2 references, sorted by distance\n2. distance_cm 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\": \"windowsill\", \"relation\": \"side_below\", \"distance_cm\": 5.6}, {\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 17.2}]}", "target": "mouse", "split": "validation"}
{"id": "fe3fc057a1_4", "scene_id": "fe3fc057a1", "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": "validation"}
{"id": "fe3fc057a1_5", "scene_id": "fe3fc057a1", "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 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 windowsill, about 6 cm away.", "target": "mug", "split": "validation"}
{"id": "fe3fc057a1_6", "scene_id": "fe3fc057a1", "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\": \"table\", \"references\": [{\"object\": \"heater\", \"relation\": \"side_above\", \"distance_cm\": 5.8}, {\"object\": \"windowsill\", \"relation\": \"side_below\", \"distance_cm\": 6.1}]}", "target": "mug", "split": "validation"}
{"id": "fe3fc057a1_7", "scene_id": "fe3fc057a1", "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": "validation"}
{"id": "fe3fc057a1_8", "scene_id": "fe3fc057a1", "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. kitchen cabinet\nB. stovetop\nC. frying pan\nD. kitchen shelf", "choices": ["A. kitchen cabinet", "B. stovetop", "C. frying pan", "D. kitchen shelf"], "answer": "B. stovetop", "target": "phone charger", "split": "validation"}
{"id": "fe3fc057a1_9", "scene_id": "fe3fc057a1", "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 stovetop, about 26 cm away. It is also to the side of and below the kitchen shelf, about 39 cm away. It is further to the side of and below the frying pan, about 39 cm away.", "target": "phone charger", "split": "validation"}
{"id": "fe3fc057a1_10", "scene_id": "fe3fc057a1", "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\": \"kitchen shelf\", \"references\": [{\"object\": \"stovetop\", \"relation\": \"side_below\", \"distance_cm\": 25.7}, {\"object\": \"kitchen shelf\", \"relation\": \"side_below\", \"distance_cm\": 39.0}]}", "target": "phone charger", "split": "validation"}