Add files using upload-large-folder tool

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000019/group_2/index_tensor([1225, 1225])_group_2_rewards.json

@@ -0,0 +1,882 @@
+[
+    {
+        "prompt": "tracking shot captured from the side, the camera tilted downward, capturing in stunning ultra-slow motion and high frame rate the sight of an adult male bald eagle swooping down like an arrow towards a calm lake illuminated by the golden morning sun. The eagle's massive, deep brown wings fold back to maximize speed as it nears the water's surface. Its pure white head lowers, its piercing golden eyes fixed on a shimmering salmon beneath the surface, its powerful yellow talons fully extended, poised for the fatal strike. As the claws plunge into the water with pinpoint precision, a magnificent crown of countless crystalline water droplets is unleashed. Caught in the extreme slow motion, each splashing droplet appears frozen in time, some forming perfect spheres, others stretched into ellipses in the air, each reflecting the warm sunlight like a suspended diamond, while the eagle's talons securely grasp its struggling prey. The whole picture, against the backdrop of the deep green pine forest and mountains in the distance, is full of primitive power and breathtaking natural beauty. It perfectly combines the dynamic moment of hunting with the static aesthetics of slow motion, creating a solemn and tense atmosphere.",
+        "seed": 29891,
+        "video_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000019/group_2/index_1225_seed_29891_dp_4_rank_8_sample_0_slot_0.mp4",
+        "sample_idx": 0,
+        "ranks_per_group": 2,
+        "group_idx": 2,
+        "rank_in_group": 0,
+        "sample_idx_in_group": 0,
+        "avg": -0.017273102328181265,
+        "vlm_logprob": -0.017273102328181265,
+        "vlm_yes_no_margin": 4.15,
+        "vlm_token_credit": -0.017273102328181265,
+        "vlm_yes_ratio": 1.0,
+        "vlm_num_questions": 5.0,
+        "reward_records": [
+            {
+                "sample_idx": 0,
+                "question_id": "q1",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.016634341329336166
+                ],
+                "opposite_logprob_tokens": [
+                    -4.141634464263916
+                ],
+                "logprob_reward": -0.016634341329336166,
+                "yes_no_margin_reward": 4.125,
+                "token_credit_reward": -0.016634341329336166,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.3544091284275055,
+                    0.10063142329454422,
+                    0.11415596306324005,
+                    0.07032905519008636,
+                    0.11067913472652435,
+                    0.029837600886821747,
+                    0.03506234288215637,
+                    0.04835028201341629,
+                    0.08210787922143936,
+                    0.054437194019556046
+                ],
+                "top_frames": [
+                    0,
+                    2,
+                    4
+                ],
+                "selected_reward": -0.016634341329336166
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q2",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.01607024483382702
+                ],
+                "opposite_logprob_tokens": [
+                    -4.141070365905762
+                ],
+                "logprob_reward": -0.01607024483382702,
+                "yes_no_margin_reward": 4.125,
+                "token_credit_reward": -0.01607024483382702,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.31755781173706055,
+                    0.09443125128746033,
+                    0.060244809836149216,
+                    0.056770794093608856,
+                    0.14383061230182648,
+                    0.047171689569950104,
+                    0.0570111908018589,
+                    0.13352100551128387,
+                    0.04878499358892441,
+                    0.04067584127187729
+                ],
+                "top_frames": [
+                    0,
+                    4,
+                    7
+                ],
+                "selected_reward": -0.01607024483382702
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q3",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.02635362558066845
+                ],
+                "opposite_logprob_tokens": [
+                    -3.651353597640991
+                ],
+                "logprob_reward": -0.02635362558066845,
+                "yes_no_margin_reward": 3.625,
+                "token_credit_reward": -0.02635362558066845,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.24749794602394104,
+                    0.08743257075548172,
+                    0.1090058907866478,
+                    0.06793955713510513,
+                    0.0535905547440052,
+                    0.04009323939681053,
+                    0.187838613986969,
+                    0.09429731220006943,
+                    0.05323851853609085,
+                    0.0590657964348793
+                ],
+                "top_frames": [
+                    0,
+                    6,
+                    2
+                ],
+                "selected_reward": -0.02635362558066845
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q4",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.020578570663928986
+                ],
+                "opposite_logprob_tokens": [
+                    -3.895578622817993
+                ],
+                "logprob_reward": -0.020578570663928986,
+                "yes_no_margin_reward": 3.875,
+                "token_credit_reward": -0.020578570663928986,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.3795311152935028,
+                    0.10192067176103592,
+                    0.08252738416194916,
+                    0.05337899923324585,
+                    0.06072967126965523,
+                    0.03404955193400383,
+                    0.10529658943414688,
+                    0.040246106684207916,
+                    0.05584465712308884,
+                    0.08647527545690536
+                ],
+                "top_frames": [
+                    0,
+                    6,
+                    1
+                ],
+                "selected_reward": -0.020578570663928986
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q5",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.006728729233145714
+                ],
+                "opposite_logprob_tokens": [
+                    -5.006728649139404
+                ],
+                "logprob_reward": -0.006728729233145714,
+                "yes_no_margin_reward": 5.0,
+                "token_credit_reward": -0.006728729233145714,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.20124807953834534,
+                    0.3744271397590637,
+                    0.03239048272371292,
+                    0.035947855561971664,
+                    0.08983442932367325,
+                    0.019620219245553017,
+                    0.03429285064339638,
+                    0.051067449152469635,
+                    0.02315940335392952,
+                    0.1380121409893036
+                ],
+                "top_frames": [
+                    1,
+                    0,
+                    9
+                ],
+                "selected_reward": -0.006728729233145714
+            }
+        ],
+        "vlm_decoded_outputs": [
+            "Yes",
+            "Yes",
+            "Yes",
+            "Yes",
+            "Yes"
+        ]
+    },
+    {
+        "prompt": "tracking shot captured from the side, the camera tilted downward, capturing in stunning ultra-slow motion and high frame rate the sight of an adult male bald eagle swooping down like an arrow towards a calm lake illuminated by the golden morning sun. The eagle's massive, deep brown wings fold back to maximize speed as it nears the water's surface. Its pure white head lowers, its piercing golden eyes fixed on a shimmering salmon beneath the surface, its powerful yellow talons fully extended, poised for the fatal strike. As the claws plunge into the water with pinpoint precision, a magnificent crown of countless crystalline water droplets is unleashed. Caught in the extreme slow motion, each splashing droplet appears frozen in time, some forming perfect spheres, others stretched into ellipses in the air, each reflecting the warm sunlight like a suspended diamond, while the eagle's talons securely grasp its struggling prey. The whole picture, against the backdrop of the deep green pine forest and mountains in the distance, is full of primitive power and breathtaking natural beauty. It perfectly combines the dynamic moment of hunting with the static aesthetics of slow motion, creating a solemn and tense atmosphere.",
+        "seed": 29892,
+        "video_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000019/group_2/index_1225_seed_29892_dp_4_rank_8_sample_1_slot_1.mp4",
+        "sample_idx": 1,
+        "ranks_per_group": 2,
+        "group_idx": 2,
+        "rank_in_group": 0,
+        "sample_idx_in_group": 1,
+        "avg": -0.02133572958409786,
+        "vlm_logprob": -0.02133572958409786,
+        "vlm_yes_no_margin": 4.05,
+        "vlm_token_credit": -0.02133572958409786,
+        "vlm_yes_ratio": 1.0,
+        "vlm_num_questions": 5.0,
+        "reward_records": [
+            {
+                "sample_idx": 0,
+                "question_id": "q1",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.03902430459856987
+                ],
+                "opposite_logprob_tokens": [
+                    -3.2890243530273438
+                ],
+                "logprob_reward": -0.03902430459856987,
+                "yes_no_margin_reward": 3.25,
+                "token_credit_reward": -0.03902430459856987,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.05568227916955948,
+                    0.15043792128562927,
+                    0.045335184782743454,
+                    0.14421117305755615,
+                    0.03527544438838959,
+                    0.14258860051631927,
+                    0.07428836822509766,
+                    0.1438307911157608,
+                    0.022798074409365654,
+                    0.18555216491222382
+                ],
+                "top_frames": [
+                    9,
+                    1,
+                    3
+                ],
+                "selected_reward": -0.03902430459856987
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q2",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.020593518391251564
+                ],
+                "opposite_logprob_tokens": [
+                    -3.8955934047698975
+                ],
+                "logprob_reward": -0.020593518391251564,
+                "yes_no_margin_reward": 3.875,
+                "token_credit_reward": -0.020593518391251564,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.017918305471539497,
+                    0.04148276522755623,
+                    0.019106481224298477,
+                    0.05399760976433754,
+                    0.20570580661296844,
+                    0.14580875635147095,
+                    0.08335360884666443,
+                    0.341793417930603,
+                    0.021786963567137718,
+                    0.0690462738275528
+                ],
+                "top_frames": [
+                    7,
+                    4,
+                    5
+                ],
+                "selected_reward": -0.020593518391251564
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q3",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.02636036090552807
+                ],
+                "opposite_logprob_tokens": [
+                    -3.651360273361206
+                ],
+                "logprob_reward": -0.02636036090552807,
+                "yes_no_margin_reward": 3.625,
+                "token_credit_reward": -0.02636036090552807,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.019345857203006744,
+                    0.04344134405255318,
+                    0.02381148561835289,
+                    0.0820787325501442,
+                    0.0352591834962368,
+                    0.13237552344799042,
+                    0.1764843910932541,
+                    0.3479873538017273,
+                    0.08785881847143173,
+                    0.051357369869947433
+                ],
+                "top_frames": [
+                    7,
+                    6,
+                    5
+                ],
+                "selected_reward": -0.02636036090552807
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q4",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.016068601980805397
+                ],
+                "opposite_logprob_tokens": [
+                    -4.141068458557129
+                ],
+                "logprob_reward": -0.016068601980805397,
+                "yes_no_margin_reward": 4.125,
+                "token_credit_reward": -0.016068601980805397,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.07570353895425797,
+                    0.11311407387256622,
+                    0.06187807768583298,
+                    0.11119620501995087,
+                    0.03153417631983757,
+                    0.11760646104812622,
+                    0.19402548670768738,
+                    0.06684552878141403,
+                    0.03226887062191963,
+                    0.19582755863666534
+                ],
+                "top_frames": [
+                    9,
+                    6,
+                    5
+                ],
+                "selected_reward": -0.016068601980805397
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q5",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.004631862044334412
+                ],
+                "opposite_logprob_tokens": [
+                    -5.379631996154785
+                ],
+                "logprob_reward": -0.004631862044334412,
+                "yes_no_margin_reward": 5.375,
+                "token_credit_reward": -0.004631862044334412,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.2302158921957016,
+                    0.12476576864719391,
+                    0.1671251356601715,
+                    0.04574928060173988,
+                    0.04850210249423981,
+                    0.06356443464756012,
+                    0.05659206956624985,
+                    0.06418290734291077,
+                    0.06868105381727219,
+                    0.13062138855457306
+                ],
+                "top_frames": [
+                    0,
+                    2,
+                    9
+                ],
+                "selected_reward": -0.004631862044334412
+            }
+        ],
+        "vlm_decoded_outputs": [
+            "Yes",
+            "Yes",
+            "Yes",
+            "Yes",
+            "Yes"
+        ]
+    },
+    {
+        "prompt": "tracking shot captured from the side, the camera tilted downward, capturing in stunning ultra-slow motion and high frame rate the sight of an adult male bald eagle swooping down like an arrow towards a calm lake illuminated by the golden morning sun. The eagle's massive, deep brown wings fold back to maximize speed as it nears the water's surface. Its pure white head lowers, its piercing golden eyes fixed on a shimmering salmon beneath the surface, its powerful yellow talons fully extended, poised for the fatal strike. As the claws plunge into the water with pinpoint precision, a magnificent crown of countless crystalline water droplets is unleashed. Caught in the extreme slow motion, each splashing droplet appears frozen in time, some forming perfect spheres, others stretched into ellipses in the air, each reflecting the warm sunlight like a suspended diamond, while the eagle's talons securely grasp its struggling prey. The whole picture, against the backdrop of the deep green pine forest and mountains in the distance, is full of primitive power and breathtaking natural beauty. It perfectly combines the dynamic moment of hunting with the static aesthetics of slow motion, creating a solemn and tense atmosphere.",
+        "seed": 29893,
+        "video_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000019/group_2/index_1225_seed_29893_dp_5_rank_10_sample_2_slot_0.mp4",
+        "sample_idx": 0,
+        "ranks_per_group": 2,
+        "group_idx": 2,
+        "rank_in_group": 1,
+        "sample_idx_in_group": 2,
+        "avg": -0.028873871732503175,
+        "vlm_logprob": -0.028873871732503175,
+        "vlm_yes_no_margin": 3.7,
+        "vlm_token_credit": -0.028873871732503175,
+        "vlm_yes_ratio": 1.0,
+        "vlm_num_questions": 5.0,
+        "reward_records": [
+            {
+                "sample_idx": 0,
+                "question_id": "q1",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.044003382325172424
+                ],
+                "opposite_logprob_tokens": [
+                    -3.169003486633301
+                ],
+                "logprob_reward": -0.044003382325172424,
+                "yes_no_margin_reward": 3.125,
+                "token_credit_reward": -0.044003382325172424,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.0748140960931778,
+                    0.1991504430770874,
+                    0.35246947407722473,
+                    0.09148301929235458,
+                    0.03986194729804993,
+                    0.038874588906764984,
+                    0.05152000114321709,
+                    0.05920584872364998,
+                    0.05377859249711037,
+                    0.038841988891363144
+                ],
+                "top_frames": [
+                    2,
+                    1,
+                    3
+                ],
+                "selected_reward": -0.044003382325172424
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q2",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.026362333446741104
+                ],
+                "opposite_logprob_tokens": [
+                    -3.651362419128418
+                ],
+                "logprob_reward": -0.026362333446741104,
+                "yes_no_margin_reward": 3.625,
+                "token_credit_reward": -0.026362333446741104,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.09110113978385925,
+                    0.12485821545124054,
+                    0.136912003159523,
+                    0.0999688059091568,
+                    0.0560699962079525,
+                    0.3296341896057129,
+                    0.028109589591622353,
+                    0.019639654085040092,
+                    0.08121924847364426,
+                    0.0324871800839901
+                ],
+                "top_frames": [
+                    5,
+                    2,
+                    1
+                ],
+                "selected_reward": -0.026362333446741104
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q3",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.04309149086475372
+                ],
+                "opposite_logprob_tokens": [
+                    -3.1680915355682373
+                ],
+                "logprob_reward": -0.04309149086475372,
+                "yes_no_margin_reward": 3.125,
+                "token_credit_reward": -0.04309149086475372,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.0842275395989418,
+                    0.19669988751411438,
+                    0.18911996483802795,
+                    0.08651993423700333,
+                    0.16582727432250977,
+                    0.04968223348259926,
+                    0.029964299872517586,
+                    0.0428336001932621,
+                    0.12737306952476501,
+                    0.027752183377742767
+                ],
+                "top_frames": [
+                    1,
+                    2,
+                    4
+                ],
+                "selected_reward": -0.04309149086475372
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q4",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.023289354518055916
+                ],
+                "opposite_logprob_tokens": [
+                    -3.773289442062378
+                ],
+                "logprob_reward": -0.023289354518055916,
+                "yes_no_margin_reward": 3.75,
+                "token_credit_reward": -0.023289354518055916,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.07553824037313461,
+                    0.34600913524627686,
+                    0.1331847757101059,
+                    0.09724350273609161,
+                    0.06454157829284668,
+                    0.04269569367170334,
+                    0.02445875108242035,
+                    0.05161132290959358,
+                    0.11687759310007095,
+                    0.04783939570188522
+                ],
+                "top_frames": [
+                    1,
+                    2,
+                    8
+                ],
+                "selected_reward": -0.023289354518055916
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q5",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.007622797507792711
+                ],
+                "opposite_logprob_tokens": [
+                    -4.882622718811035
+                ],
+                "logprob_reward": -0.007622797507792711,
+                "yes_no_margin_reward": 4.875,
+                "token_credit_reward": -0.007622797507792711,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.046300992369651794,
+                    0.19635072350502014,
+                    0.16885940730571747,
+                    0.3830609619617462,
+                    0.027374811470508575,
+                    0.06821209192276001,
+                    0.04533887654542923,
+                    0.027184361591935158,
+                    0.015993235632777214,
+                    0.02132454700767994
+                ],
+                "top_frames": [
+                    3,
+                    1,
+                    2
+                ],
+                "selected_reward": -0.007622797507792711
+            }
+        ],
+        "vlm_decoded_outputs": [
+            "Yes",
+            "Yes",
+            "Yes",
+            "Yes",
+            "Yes"
+        ]
+    },
+    {
+        "prompt": "tracking shot captured from the side, the camera tilted downward, capturing in stunning ultra-slow motion and high frame rate the sight of an adult male bald eagle swooping down like an arrow towards a calm lake illuminated by the golden morning sun. The eagle's massive, deep brown wings fold back to maximize speed as it nears the water's surface. Its pure white head lowers, its piercing golden eyes fixed on a shimmering salmon beneath the surface, its powerful yellow talons fully extended, poised for the fatal strike. As the claws plunge into the water with pinpoint precision, a magnificent crown of countless crystalline water droplets is unleashed. Caught in the extreme slow motion, each splashing droplet appears frozen in time, some forming perfect spheres, others stretched into ellipses in the air, each reflecting the warm sunlight like a suspended diamond, while the eagle's talons securely grasp its struggling prey. The whole picture, against the backdrop of the deep green pine forest and mountains in the distance, is full of primitive power and breathtaking natural beauty. It perfectly combines the dynamic moment of hunting with the static aesthetics of slow motion, creating a solemn and tense atmosphere.",
+        "seed": 29894,
+        "video_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000019/group_2/index_1225_seed_29894_dp_5_rank_10_sample_3_slot_1.mp4",
+        "sample_idx": 1,
+        "ranks_per_group": 2,
+        "group_idx": 2,
+        "rank_in_group": 1,
+        "sample_idx_in_group": 3,
+        "avg": -0.0275817783549428,
+        "vlm_logprob": -0.0275817783549428,
+        "vlm_yes_no_margin": 3.8,
+        "vlm_token_credit": -0.0275817783549428,
+        "vlm_yes_ratio": 1.0,
+        "vlm_num_questions": 5.0,
+        "reward_records": [
+            {
+                "sample_idx": 0,
+                "question_id": "q1",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.06286749988794327
+                ],
+                "opposite_logprob_tokens": [
+                    -2.8128674030303955
+                ],
+                "logprob_reward": -0.06286749988794327,
+                "yes_no_margin_reward": 2.75,
+                "token_credit_reward": -0.06286749988794327,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.4594794809818268,
+                    0.04368318244814873,
+                    0.060214102268218994,
+                    0.027757829055190086,
+                    0.03648347780108452,
+                    0.03647899255156517,
+                    0.023013755679130554,
+                    0.02014220505952835,
+                    0.27039942145347595,
+                    0.022347552701830864
+                ],
+                "top_frames": [
+                    0,
+                    8,
+                    2
+                ],
+                "selected_reward": -0.06286749988794327
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q2",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.026357226073741913
+                ],
+                "opposite_logprob_tokens": [
+                    -3.6513571739196777
+                ],
+                "logprob_reward": -0.026357226073741913,
+                "yes_no_margin_reward": 3.625,
+                "token_credit_reward": -0.026357226073741913,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.2637869715690613,
+                    0.05487702786922455,
+                    0.2610485851764679,
+                    0.03293614834547043,
+                    0.05391418933868408,
+                    0.10205941647291183,
+                    0.048432137817144394,
+                    0.03173154219985008,
+                    0.06106520816683769,
+                    0.09014879167079926
+                ],
+                "top_frames": [
+                    0,
+                    2,
+                    5
+                ],
+                "selected_reward": -0.026357226073741913
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q3",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.02636883594095707
+                ],
+                "opposite_logprob_tokens": [
+                    -3.6513688564300537
+                ],
+                "logprob_reward": -0.02636883594095707,
+                "yes_no_margin_reward": 3.625,
+                "token_credit_reward": -0.02636883594095707,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.19347023963928223,
+                    0.05968620255589485,
+                    0.16033679246902466,
+                    0.042797379195690155,
+                    0.06923689693212509,
+                    0.2502368688583374,
+                    0.02104455791413784,
+                    0.014521404169499874,
+                    0.12968094646930695,
+                    0.05898871272802353
+                ],
+                "top_frames": [
+                    5,
+                    0,
+                    2
+                ],
+                "selected_reward": -0.02636883594095707
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q4",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.01254053320735693
+                ],
+                "opposite_logprob_tokens": [
+                    -4.387540340423584
+                ],
+                "logprob_reward": -0.01254053320735693,
+                "yes_no_margin_reward": 4.375,
+                "token_credit_reward": -0.01254053320735693,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.3020077645778656,
+                    0.056703101843595505,
+                    0.10533630847930908,
+                    0.04649340361356735,
+                    0.04999338835477829,
+                    0.24498365819454193,
+                    0.05534031614661217,
+                    0.043337296694517136,
+                    0.0636034682393074,
+                    0.032201334834098816
+                ],
+                "top_frames": [
+                    0,
+                    5,
+                    2
+                ],
+                "selected_reward": -0.01254053320735693
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q5",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the main subject an adult male bald eagle with a pure white head and deep brown wings?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the eagle performing a swooping motion towards a calm lake while its wings fold back?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Are the eagle's yellow talons fully extended and grasping a struggling salmon beneath the water's surface?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Is the scene captured in ultra-slow motion showing water droplets frozen in time as spheres or ellipses reflecting sunlight?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the background composed of a deep green pine forest with mountains in the distance?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.009774796664714813
+                ],
+                "opposite_logprob_tokens": [
+                    -4.634774684906006
+                ],
+                "logprob_reward": -0.009774796664714813,
+                "yes_no_margin_reward": 4.625,
+                "token_credit_reward": -0.009774796664714813,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.21794357895851135,
+                    0.04974507540464401,
+                    0.3961476683616638,
+                    0.128492534160614,
+                    0.050274014472961426,
+                    0.02622945047914982,
+                    0.026263372972607613,
+                    0.02754315733909607,
+                    0.025207925587892532,
+                    0.05215319246053696
+                ],
+                "top_frames": [
+                    2,
+                    0,
+                    3
+                ],
+                "selected_reward": -0.009774796664714813
+            }
+        ],
+        "vlm_decoded_outputs": [
+            "Yes",
+            "Yes",
+            "Yes",
+            "Yes",
+            "Yes"
+        ]
+    }
+]

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000019/group_5/index_tensor([4475, 4475])_group_5_rewards.json

@@ -0,0 +1,882 @@
+[
+    {
+        "prompt": "An incredibly close, static, aerial view captures the moment a thick, smooth mass of pure white heavy cream drops into the center of a black ceramic cup filled with steaming black coffee. The camera remains perfectly still, focused on the liquid's dark, reflective surface. As the cream makes contact, it dives beneath the surface and immediately begins to bloom upward and outward in a slow, mesmerizing fashion. Billowing plumes of opaque white unfold like cosmic clouds, creating intricate, mesmerizing swirls and delicate creamy tendrils that dance and twist within the deep brown liquid. The stark contrast between the bright white coffee and the dark coffee slowly softens as they blend, producing a beautiful, evolving swirling pattern of creamy beige and caramel. The entire process is illuminated by a soft, directional light, casting a gentle sheen on the rotating surface, creating a tranquil, hypnotic, and visually rich atmosphere focused on the beautiful, chaotic ballet of two liquids becoming one.",
+        "seed": 13153,
+        "video_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000019/group_5/index_4475_seed_13153_dp_10_rank_20_sample_0_slot_0.mp4",
+        "sample_idx": 0,
+        "ranks_per_group": 2,
+        "group_idx": 5,
+        "rank_in_group": 0,
+        "sample_idx_in_group": 0,
+        "avg": -0.27226154222153126,
+        "vlm_logprob": -0.27226154222153126,
+        "vlm_yes_no_margin": 3.075000011920929,
+        "vlm_token_credit": -0.27226154222153126,
+        "vlm_yes_ratio": 0.8,
+        "vlm_num_questions": 5.0,
+        "reward_records": [
+            {
+                "sample_idx": 0,
+                "question_id": "q1",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.1037372574210167
+                ],
+                "opposite_logprob_tokens": [
+                    -2.3537373542785645
+                ],
+                "logprob_reward": -0.1037372574210167,
+                "yes_no_margin_reward": 2.25,
+                "token_credit_reward": -0.1037372574210167,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.07030411809682846,
+                    0.48889780044555664,
+                    0.062220655381679535,
+                    0.03784313425421715,
+                    0.12675191462039948,
+                    0.04559764638543129,
+                    0.03533904254436493,
+                    0.035155702382326126,
+                    0.07989601045846939,
+                    0.017993975430727005
+                ],
+                "top_frames": [
+                    1,
+                    4,
+                    8
+                ],
+                "selected_reward": -0.1037372574210167
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q2",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "No",
+                "normalized_pred_keywords": [
+                    "No"
+                ],
+                "target_logprob_tokens": [
+                    -1.2236621379852295
+                ],
+                "opposite_logprob_tokens": [
+                    -0.34866219758987427
+                ],
+                "logprob_reward": -1.2236621379852295,
+                "yes_no_margin_reward": -0.8749999403953552,
+                "token_credit_reward": -1.2236621379852295,
+                "yes_ratio_reward": 0.0,
+                "frame_importance": [
+                    0.07414276152849197,
+                    0.06302167475223541,
+                    0.13199889659881592,
+                    0.19615323841571808,
+                    0.12242792546749115,
+                    0.10030072927474976,
+                    0.06037510186433792,
+                    0.12629815936088562,
+                    0.08081300556659698,
+                    0.04446852207183838
+                ],
+                "top_frames": [
+                    3,
+                    2,
+                    7
+                ],
+                "selected_reward": -1.2236621379852295
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q3",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.0036272702272981405
+                ],
+                "opposite_logprob_tokens": [
+                    -5.628627300262451
+                ],
+                "logprob_reward": -0.0036272702272981405,
+                "yes_no_margin_reward": 5.625,
+                "token_credit_reward": -0.0036272702272981405,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.0543455071747303,
+                    0.11673092097043991,
+                    0.05714545398950577,
+                    0.15274186432361603,
+                    0.035377055406570435,
+                    0.08092499524354935,
+                    0.09107499569654465,
+                    0.1685304045677185,
+                    0.22699914872646332,
+                    0.016129666939377785
+                ],
+                "top_frames": [
+                    8,
+                    7,
+                    3
+                ],
+                "selected_reward": -0.0036272702272981405
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q4",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.01420364435762167
+                ],
+                "opposite_logprob_tokens": [
+                    -4.2642035484313965
+                ],
+                "logprob_reward": -0.01420364435762167,
+                "yes_no_margin_reward": 4.25,
+                "token_credit_reward": -0.01420364435762167,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.054000210016965866,
+                    0.10132480412721634,
+                    0.05161728709936142,
+                    0.09289602190256119,
+                    0.06855905801057816,
+                    0.11097538471221924,
+                    0.09169010818004608,
+                    0.08433806896209717,
+                    0.3015536963939667,
+                    0.04304538667201996
+                ],
+                "top_frames": [
+                    8,
+                    5,
+                    1
+                ],
+                "selected_reward": -0.01420364435762167
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q5",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.016077401116490364
+                ],
+                "opposite_logprob_tokens": [
+                    -4.141077518463135
+                ],
+                "logprob_reward": -0.016077401116490364,
+                "yes_no_margin_reward": 4.125,
+                "token_credit_reward": -0.016077401116490364,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.11808724701404572,
+                    0.10579265654087067,
+                    0.1433478593826294,
+                    0.08125042170286179,
+                    0.10432972013950348,
+                    0.1630369871854782,
+                    0.06680206954479218,
+                    0.08857079595327377,
+                    0.09323268383741379,
+                    0.0355495847761631
+                ],
+                "top_frames": [
+                    5,
+                    2,
+                    0
+                ],
+                "selected_reward": -0.016077401116490364
+            }
+        ],
+        "vlm_decoded_outputs": [
+            "Yes",
+            "No",
+            "Yes",
+            "Yes",
+            "Yes"
+        ]
+    },
+    {
+        "prompt": "An incredibly close, static, aerial view captures the moment a thick, smooth mass of pure white heavy cream drops into the center of a black ceramic cup filled with steaming black coffee. The camera remains perfectly still, focused on the liquid's dark, reflective surface. As the cream makes contact, it dives beneath the surface and immediately begins to bloom upward and outward in a slow, mesmerizing fashion. Billowing plumes of opaque white unfold like cosmic clouds, creating intricate, mesmerizing swirls and delicate creamy tendrils that dance and twist within the deep brown liquid. The stark contrast between the bright white coffee and the dark coffee slowly softens as they blend, producing a beautiful, evolving swirling pattern of creamy beige and caramel. The entire process is illuminated by a soft, directional light, casting a gentle sheen on the rotating surface, creating a tranquil, hypnotic, and visually rich atmosphere focused on the beautiful, chaotic ballet of two liquids becoming one.",
+        "seed": 13154,
+        "video_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000019/group_5/index_4475_seed_13154_dp_10_rank_20_sample_1_slot_1.mp4",
+        "sample_idx": 1,
+        "ranks_per_group": 2,
+        "group_idx": 5,
+        "rank_in_group": 0,
+        "sample_idx_in_group": 1,
+        "avg": -0.2442528204061091,
+        "vlm_logprob": -0.2442528204061091,
+        "vlm_yes_no_margin": 3.15,
+        "vlm_token_credit": -0.2442528204061091,
+        "vlm_yes_ratio": 0.8,
+        "vlm_num_questions": 5.0,
+        "reward_records": [
+            {
+                "sample_idx": 0,
+                "question_id": "q1",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.04664944112300873
+                ],
+                "opposite_logprob_tokens": [
+                    -3.17164945602417
+                ],
+                "logprob_reward": -0.04664944112300873,
+                "yes_no_margin_reward": 3.125,
+                "token_credit_reward": -0.04664944112300873,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.12578918039798737,
+                    0.06934741884469986,
+                    0.577904224395752,
+                    0.09869707375764847,
+                    0.014369872398674488,
+                    0.012793570756912231,
+                    0.01970050297677517,
+                    0.03865700587630272,
+                    0.023121409118175507,
+                    0.019619692116975784
+                ],
+                "top_frames": [
+                    2,
+                    0,
+                    3
+                ],
+                "selected_reward": -0.04664944112300873
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q2",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "No",
+                "normalized_pred_keywords": [
+                    "No"
+                ],
+                "target_logprob_tokens": [
+                    -1.137110710144043
+                ],
+                "opposite_logprob_tokens": [
+                    -0.38711073994636536
+                ],
+                "logprob_reward": -1.137110710144043,
+                "yes_no_margin_reward": -0.75,
+                "token_credit_reward": -1.137110710144043,
+                "yes_ratio_reward": 0.0,
+                "frame_importance": [
+                    0.17486834526062012,
+                    0.13457255065441132,
+                    0.16217878460884094,
+                    0.15074722468852997,
+                    0.019550597295165062,
+                    0.04974162578582764,
+                    0.15348894894123077,
+                    0.03757468983530998,
+                    0.024755464866757393,
+                    0.0925218015909195
+                ],
+                "top_frames": [
+                    0,
+                    2,
+                    6
+                ],
+                "selected_reward": -1.137110710144043
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q3",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.006753476802259684
+                ],
+                "opposite_logprob_tokens": [
+                    -5.006753444671631
+                ],
+                "logprob_reward": -0.006753476802259684,
+                "yes_no_margin_reward": 5.0,
+                "token_credit_reward": -0.006753476802259684,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.13349223136901855,
+                    0.058299869298934937,
+                    0.19995801150798798,
+                    0.08435423672199249,
+                    0.057562436908483505,
+                    0.17024624347686768,
+                    0.13459590077400208,
+                    0.08704138547182083,
+                    0.036989327520132065,
+                    0.03746037185192108
+                ],
+                "top_frames": [
+                    2,
+                    5,
+                    6
+                ],
+                "selected_reward": -0.006753476802259684
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q4",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.018202170729637146
+                ],
+                "opposite_logprob_tokens": [
+                    -4.018202304840088
+                ],
+                "logprob_reward": -0.018202170729637146,
+                "yes_no_margin_reward": 4.0,
+                "token_credit_reward": -0.018202170729637146,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.027596300467848778,
+                    0.019701816141605377,
+                    0.6981757879257202,
+                    0.046480942517519,
+                    0.02039325051009655,
+                    0.030135320499539375,
+                    0.11256333440542221,
+                    0.018896497786045074,
+                    0.011023981496691704,
+                    0.015032785013318062
+                ],
+                "top_frames": [
+                    2,
+                    6,
+                    3
+                ],
+                "selected_reward": -0.018202170729637146
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q5",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.012548303231596947
+                ],
+                "opposite_logprob_tokens": [
+                    -4.387548446655273
+                ],
+                "logprob_reward": -0.012548303231596947,
+                "yes_no_margin_reward": 4.375,
+                "token_credit_reward": -0.012548303231596947,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.10590685158967972,
+                    0.027642671018838882,
+                    0.6347907781600952,
+                    0.06965061277151108,
+                    0.01592395454645157,
+                    0.027257690206170082,
+                    0.03625432774424553,
+                    0.030835822224617004,
+                    0.03039637766778469,
+                    0.021340837702155113
+                ],
+                "top_frames": [
+                    2,
+                    0,
+                    3
+                ],
+                "selected_reward": -0.012548303231596947
+            }
+        ],
+        "vlm_decoded_outputs": [
+            "Yes",
+            "No",
+            "Yes",
+            "Yes",
+            "Yes"
+        ]
+    },
+    {
+        "prompt": "An incredibly close, static, aerial view captures the moment a thick, smooth mass of pure white heavy cream drops into the center of a black ceramic cup filled with steaming black coffee. The camera remains perfectly still, focused on the liquid's dark, reflective surface. As the cream makes contact, it dives beneath the surface and immediately begins to bloom upward and outward in a slow, mesmerizing fashion. Billowing plumes of opaque white unfold like cosmic clouds, creating intricate, mesmerizing swirls and delicate creamy tendrils that dance and twist within the deep brown liquid. The stark contrast between the bright white coffee and the dark coffee slowly softens as they blend, producing a beautiful, evolving swirling pattern of creamy beige and caramel. The entire process is illuminated by a soft, directional light, casting a gentle sheen on the rotating surface, creating a tranquil, hypnotic, and visually rich atmosphere focused on the beautiful, chaotic ballet of two liquids becoming one.",
+        "seed": 13155,
+        "video_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000019/group_5/index_4475_seed_13155_dp_11_rank_22_sample_2_slot_0.mp4",
+        "sample_idx": 0,
+        "ranks_per_group": 2,
+        "group_idx": 5,
+        "rank_in_group": 1,
+        "sample_idx_in_group": 2,
+        "avg": -0.28215617099776863,
+        "vlm_logprob": -0.28215617099776863,
+        "vlm_yes_no_margin": 3.275,
+        "vlm_token_credit": -0.28215617099776863,
+        "vlm_yes_ratio": 0.8,
+        "vlm_num_questions": 5.0,
+        "reward_records": [
+            {
+                "sample_idx": 0,
+                "question_id": "q1",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.07349292933940887
+                ],
+                "opposite_logprob_tokens": [
+                    -2.698493003845215
+                ],
+                "logprob_reward": -0.07349292933940887,
+                "yes_no_margin_reward": 2.625,
+                "token_credit_reward": -0.07349292933940887,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.10152854770421982,
+                    0.021715473383665085,
+                    0.010391980409622192,
+                    0.026708584278821945,
+                    0.7150031328201294,
+                    0.019752908498048782,
+                    0.027928832918405533,
+                    0.0403563529253006,
+                    0.0061722612008452415,
+                    0.0304418932646513
+                ],
+                "top_frames": [
+                    4,
+                    0,
+                    7
+                ],
+                "selected_reward": -0.07349292933940887
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q2",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "No",
+                "normalized_pred_keywords": [
+                    "No"
+                ],
+                "target_logprob_tokens": [
+                    -1.313459038734436
+                ],
+                "opposite_logprob_tokens": [
+                    -0.31345900893211365
+                ],
+                "logprob_reward": -1.313459038734436,
+                "yes_no_margin_reward": -1.0,
+                "token_credit_reward": -1.313459038734436,
+                "yes_ratio_reward": 0.0,
+                "frame_importance": [
+                    0.05737944692373276,
+                    0.037879686802625656,
+                    0.05862542986869812,
+                    0.04654702916741371,
+                    0.3613207936286926,
+                    0.04117792844772339,
+                    0.05106215551495552,
+                    0.1770876944065094,
+                    0.021746881306171417,
+                    0.14717291295528412
+                ],
+                "top_frames": [
+                    4,
+                    7,
+                    9
+                ],
+                "selected_reward": -1.313459038734436
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q3",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.004107966553419828
+                ],
+                "opposite_logprob_tokens": [
+                    -5.50410795211792
+                ],
+                "logprob_reward": -0.004107966553419828,
+                "yes_no_margin_reward": 5.5,
+                "token_credit_reward": -0.004107966553419828,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.1071074977517128,
+                    0.05905218794941902,
+                    0.08098366111516953,
+                    0.07996778935194016,
+                    0.21482807397842407,
+                    0.09909746795892715,
+                    0.17637746036052704,
+                    0.12667709589004517,
+                    0.021959079429507256,
+                    0.033949680626392365
+                ],
+                "top_frames": [
+                    4,
+                    6,
+                    7
+                ],
+                "selected_reward": -0.004107966553419828
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q4",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.011081461794674397
+                ],
+                "opposite_logprob_tokens": [
+                    -4.511081695556641
+                ],
+                "logprob_reward": -0.011081461794674397,
+                "yes_no_margin_reward": 4.5,
+                "token_credit_reward": -0.011081461794674397,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.11776763945817947,
+                    0.06954246759414673,
+                    0.058890312910079956,
+                    0.06410576403141022,
+                    0.21239866316318512,
+                    0.059807706624269485,
+                    0.11244271695613861,
+                    0.1802208572626114,
+                    0.0662713497877121,
+                    0.058552537113428116
+                ],
+                "top_frames": [
+                    4,
+                    7,
+                    0
+                ],
+                "selected_reward": -0.011081461794674397
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q5",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.008639458566904068
+                ],
+                "opposite_logprob_tokens": [
+                    -4.758639335632324
+                ],
+                "logprob_reward": -0.008639458566904068,
+                "yes_no_margin_reward": 4.75,
+                "token_credit_reward": -0.008639458566904068,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.13227960467338562,
+                    0.05001925304532051,
+                    0.0607333742082119,
+                    0.09916170686483383,
+                    0.11167429387569427,
+                    0.059779126197099686,
+                    0.10865134745836258,
+                    0.20157133042812347,
+                    0.06380307674407959,
+                    0.11232681572437286
+                ],
+                "top_frames": [
+                    7,
+                    0,
+                    9
+                ],
+                "selected_reward": -0.008639458566904068
+            }
+        ],
+        "vlm_decoded_outputs": [
+            "Yes",
+            "No",
+            "Yes",
+            "Yes",
+            "Yes"
+        ]
+    },
+    {
+        "prompt": "An incredibly close, static, aerial view captures the moment a thick, smooth mass of pure white heavy cream drops into the center of a black ceramic cup filled with steaming black coffee. The camera remains perfectly still, focused on the liquid's dark, reflective surface. As the cream makes contact, it dives beneath the surface and immediately begins to bloom upward and outward in a slow, mesmerizing fashion. Billowing plumes of opaque white unfold like cosmic clouds, creating intricate, mesmerizing swirls and delicate creamy tendrils that dance and twist within the deep brown liquid. The stark contrast between the bright white coffee and the dark coffee slowly softens as they blend, producing a beautiful, evolving swirling pattern of creamy beige and caramel. The entire process is illuminated by a soft, directional light, casting a gentle sheen on the rotating surface, creating a tranquil, hypnotic, and visually rich atmosphere focused on the beautiful, chaotic ballet of two liquids becoming one.",
+        "seed": 13156,
+        "video_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000019/group_5/index_4475_seed_13156_dp_11_rank_22_sample_3_slot_1.mp4",
+        "sample_idx": 1,
+        "ranks_per_group": 2,
+        "group_idx": 5,
+        "rank_in_group": 1,
+        "sample_idx_in_group": 3,
+        "avg": -0.18193239429965616,
+        "vlm_logprob": -0.18193239429965616,
+        "vlm_yes_no_margin": 3.4,
+        "vlm_token_credit": -0.18193239429965616,
+        "vlm_yes_ratio": 0.8,
+        "vlm_num_questions": 5.0,
+        "reward_records": [
+            {
+                "sample_idx": 0,
+                "question_id": "q1",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.1319090723991394
+                ],
+                "opposite_logprob_tokens": [
+                    -2.131909132003784
+                ],
+                "logprob_reward": -0.1319090723991394,
+                "yes_no_margin_reward": 2.0,
+                "token_credit_reward": -0.1319090723991394,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.2050042748451233,
+                    0.10228586196899414,
+                    0.08234160393476486,
+                    0.30630338191986084,
+                    0.20609498023986816,
+                    0.009660827927291393,
+                    0.032808661460876465,
+                    0.025457004085183144,
+                    0.015549811534583569,
+                    0.014493661001324654
+                ],
+                "top_frames": [
+                    3,
+                    4,
+                    0
+                ],
+                "selected_reward": -0.1319090723991394
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q2",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "No",
+                "normalized_pred_keywords": [
+                    "No"
+                ],
+                "target_logprob_tokens": [
+                    -0.7578462958335876
+                ],
+                "opposite_logprob_tokens": [
+                    -0.6328462958335876
+                ],
+                "logprob_reward": -0.7578462958335876,
+                "yes_no_margin_reward": -0.125,
+                "token_credit_reward": -0.7578462958335876,
+                "yes_ratio_reward": 0.0,
+                "frame_importance": [
+                    0.061527617275714874,
+                    0.37170007824897766,
+                    0.031464118510484695,
+                    0.1396683305501938,
+                    0.07987332344055176,
+                    0.06148332357406616,
+                    0.0863114669919014,
+                    0.09141837805509567,
+                    0.045191820710897446,
+                    0.03136151656508446
+                ],
+                "top_frames": [
+                    1,
+                    3,
+                    7
+                ],
+                "selected_reward": -0.7578462958335876
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q3",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.004646338522434235
+                ],
+                "opposite_logprob_tokens": [
+                    -5.379646301269531
+                ],
+                "logprob_reward": -0.004646338522434235,
+                "yes_no_margin_reward": 5.375,
+                "token_credit_reward": -0.004646338522434235,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.217682346701622,
+                    0.056492097675800323,
+                    0.057653721421957016,
+                    0.3780219554901123,
+                    0.12741774320602417,
+                    0.06109175086021423,
+                    0.012199032120406628,
+                    0.042819779366254807,
+                    0.035205379128456116,
+                    0.011416199617087841
+                ],
+                "top_frames": [
+                    3,
+                    0,
+                    4
+                ],
+                "selected_reward": -0.004646338522434235
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q4",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.007632025051862001
+                ],
+                "opposite_logprob_tokens": [
+                    -4.882632255554199
+                ],
+                "logprob_reward": -0.007632025051862001,
+                "yes_no_margin_reward": 4.875,
+                "token_credit_reward": -0.007632025051862001,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.1754501610994339,
+                    0.12800127267837524,
+                    0.09879304468631744,
+                    0.06135519593954086,
+                    0.08531349152326584,
+                    0.06916338950395584,
+                    0.21562470495700836,
+                    0.07760372012853622,
+                    0.039387039840221405,
+                    0.04930797591805458
+                ],
+                "top_frames": [
+                    6,
+                    0,
+                    1
+                ],
+                "selected_reward": -0.007632025051862001
+            },
+            {
+                "sample_idx": 0,
+                "question_id": "q5",
+                "question": "You are a careful video question-answering assistant.\nUse only visually grounded evidence from the video, and do not guess unseen details.\n\nGiven an AI generated video, answer the following atomic visual questions using only evidence from the video.\n\nRules:\n- Answer each question independently.\n- Use concise canonical phrases.\n- Do not add explanations.\n- For \"yes_no\" questions, answer with \"Yes\" or \"No\" only.\n- Return valid JSON only.\n\nOutput format:\n{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"...\"}]}\n\nSub-questions:\n- {\"id\": \"q1\", \"question\": \"Is the camera view an aerial shot looking down at a black ceramic cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q2\", \"question\": \"Is the camera completely static and does not move or pan during the video?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q3\", \"question\": \"Does a thick, smooth mass of pure white heavy cream drop into the center of the cup?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q4\", \"question\": \"Does the cream dive beneath the coffee surface and bloom upward and outward in slow, billowing plumes?\", \"answer_type\": \"yes_no\"}\n- {\"id\": \"q5\", \"question\": \"Is the scene illuminated by soft, directional light casting a gentle sheen on the liquid surface?\", \"answer_type\": \"yes_no\"}",
+                "ref_answer": "{\"sub_answers\": [{\"id\": \"q1\", \"answer\": \"Yes\"}, {\"id\": \"q2\", \"answer\": \"Yes\"}, {\"id\": \"q3\", \"answer\": \"Yes\"}, {\"id\": \"q4\", \"answer\": \"Yes\"}, {\"id\": \"q5\", \"answer\": \"Yes\"}]}",
+                "target_answer": "Yes",
+                "pred_sentence": "Yes",
+                "normalized_pred_keywords": [
+                    "Yes"
+                ],
+                "target_logprob_tokens": [
+                    -0.007628239691257477
+                ],
+                "opposite_logprob_tokens": [
+                    -4.882628440856934
+                ],
+                "logprob_reward": -0.007628239691257477,
+                "yes_no_margin_reward": 4.875,
+                "token_credit_reward": -0.007628239691257477,
+                "yes_ratio_reward": 1.0,
+                "frame_importance": [
+                    0.1293148696422577,
+                    0.13911668956279755,
+                    0.066315196454525,
+                    0.24705514311790466,
+                    0.1318628489971161,
+                    0.03409482538700104,
+                    0.051732998341321945,
+                    0.14158552885055542,
+                    0.024844424799084663,
+                    0.034077391028404236
+                ],
+                "top_frames": [
+                    3,
+                    7,
+                    1
+                ],
+                "selected_reward": -0.007628239691257477
+            }
+        ],
+        "vlm_decoded_outputs": [
+            "Yes",
+            "No",
+            "Yes",
+            "Yes",
+            "Yes"
+        ]
+    }
+]

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000172/group_0/index_8254_seed_31702_dp_0_rank_0_sample_0_slot_0.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c4597859e786a2e349a5d561e78fc360ca599c195f333b4d37818adac8a53a3
+size 404400

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000172/group_0/index_8254_seed_31703_dp_0_rank_0_sample_1_slot_1.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc4524a8357e4b9c0191b0dd8536fea7135706d48440812889a81b02d0efdc2a
+size 467236

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000172/group_0/index_8254_seed_31704_dp_1_rank_2_sample_2_slot_0.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f578d085119462cb006e1f804defbcc72ee28d00c2afb6ab736a95adf3f40886
+size 463148

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000172/group_0/index_8254_seed_31705_dp_1_rank_2_sample_3_slot_1.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0bbfcfebd097bdd758901acd3315f1a79e4878124185cc4a3334be5fbc9f850
+size 430345

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000172/group_10/index_5956_seed_8420_dp_20_rank_40_sample_0_slot_0.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94d96cd68e3c404a01a01b526aa056e34f0f37f763b6909c8bf55ce6e1262da8
+size 368099

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000172/group_10/index_5956_seed_8421_dp_20_rank_40_sample_1_slot_1.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e29001aa380af942b993ff48090e69727b5b9964b1cbd1732e000c54e5b9151
+size 347693

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000172/group_10/index_5956_seed_8422_dp_21_rank_42_sample_2_slot_0.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91ee3e4d1c2ceec8bc034329623fa14b7e07e92f2323582932188a6a2e6da039
+size 226263

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000172/group_10/index_5956_seed_8423_dp_21_rank_42_sample_3_slot_1.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e8bd2f25ab0369691e701826202066234a906cbfab3197b60fb635e125d64be
+size 229471

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000172/group_15/index_7440_seed_30989_dp_30_rank_60_sample_0_slot_0.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:abe15adbed77d6491b94332a172a211a28249b11fd57d0fc054ccf5af834c9e8
+size 451555

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000172/group_15/index_7440_seed_30990_dp_30_rank_60_sample_1_slot_1.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ddcf19a7bb8dd0d44fd7ff8ceac1bb7369067177f9a94e5322da6d6219f9ea1
+size 492413

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000172/group_15/index_7440_seed_30991_dp_31_rank_62_sample_2_slot_0.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d4325dfd25aeedcb5338574e2af33960c6d50e6073d87b184f0a3b9f33ed046
+size 341561

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000172/group_15/index_7440_seed_30992_dp_31_rank_62_sample_3_slot_1.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cdf3e19e4e48effae983d17e9ba568402ea5c40eb7daae49b0c89ec454b19452
+size 455555

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000172/group_2/index_9557_seed_29456_dp_4_rank_8_sample_0_slot_0.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66b285d679866aeed0c15756a7da329b8a26ed12be303d9fa297d96cfe8fb5d0
+size 242763

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000172/group_2/index_9557_seed_29458_dp_5_rank_10_sample_2_slot_0.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30217286f6301c4c3b376f99db14561b652debb9dae2e61149cdd85bd435b41f
+size 238506

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000172/group_7/index_9598_seed_3326_dp_14_rank_28_sample_0_slot_0.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:173a63ba134871534d92527b72ac3afa1f6552f83448da9ac41d0e9a8a21d918
+size 573548

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/rl_samples/0000172/group_7/index_9598_seed_3327_dp_14_rank_28_sample_1_slot_1.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff4ab1e43dc2432c978dfb4f38f979a5cce5d90035a9c8e40ffc37e2d5fa7ba6
+size 477830

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_000_seed_310_A crystal-clear glass ball begins its journey acro_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T06:29:36.241127",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_000_seed_310_A crystal-clear glass ball begins its journey acro.mp4",
+  "arguments": {
+    "prompt": "A crystal-clear glass ball begins its journey across a polished wooden tabletop, reflecting the ambient light and casting delicate shadows. The camera remains fixed, capturing the ball's gentle, mesmerizing roll. As it moves, the ball's surface mirrors the room's soft hues and subtle textures, creating a dynamic interplay of light and shadow. The tabletop's smooth grain enhances the ball's seamless glide, while the ambient sounds of a quiet room add to the tranquil atmosphere. The ball's path is unpredictable yet graceful, embodying a serene dance of motion and stillness in a harmonious setting.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 310,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_000_seed_310_A crystal-clear glass ball begins its journey acro.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_000_seed_42_A crystal-clear glass ball begins its journey acro_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T03:02:05.902190",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_000_seed_42_A crystal-clear glass ball begins its journey acro.mp4",
+  "arguments": {
+    "prompt": "A crystal-clear glass ball begins its journey across a polished wooden tabletop, reflecting the ambient light and casting delicate shadows. The camera remains fixed, capturing the ball's gentle, mesmerizing roll. As it moves, the ball's surface mirrors the room's soft hues and subtle textures, creating a dynamic interplay of light and shadow. The tabletop's smooth grain enhances the ball's seamless glide, while the ambient sounds of a quiet room add to the tranquil atmosphere. The ball's path is unpredictable yet graceful, embodying a serene dance of motion and stillness in a harmonious setting.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 42,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_000_seed_42_A crystal-clear glass ball begins its journey acro.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_000_seed_930_A crystal-clear glass ball begins its journey acro_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-13T23:34:55.984770",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_000_seed_930_A crystal-clear glass ball begins its journey acro.mp4",
+  "arguments": {
+    "prompt": "A crystal-clear glass ball begins its journey across a polished wooden tabletop, reflecting the ambient light and casting delicate shadows. The camera remains fixed, capturing the ball's gentle, mesmerizing roll. As it moves, the ball's surface mirrors the room's soft hues and subtle textures, creating a dynamic interplay of light and shadow. The tabletop's smooth grain enhances the ball's seamless glide, while the ambient sounds of a quiet room add to the tranquil atmosphere. The ball's path is unpredictable yet graceful, embodying a serene dance of motion and stillness in a harmonious setting.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 930,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_000_seed_930_A crystal-clear glass ball begins its journey acro.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_001_seed_310_A polished metal ball gleaming under soft ambient_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T06:31:35.108601",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_001_seed_310_A polished metal ball gleaming under soft ambient.mp4",
+  "arguments": {
+    "prompt": "A polished metal ball, gleaming under soft ambient lighting, begins its journey across a sleek, dark wooden tabletop. The camera remains fixed, capturing the ball's smooth, deliberate motion as it glides effortlessly, reflecting the room's subtle hues. The tabletop's surface, impeccably smooth, allows the ball to maintain a steady, unbroken path, creating a mesmerizing dance of light and shadow. As the ball rolls, its metallic surface mirrors the surroundings, offering fleeting glimpses of the room's decor. The gentle sound of the ball's movement adds a rhythmic element, enhancing the serene, captivating atmosphere of the scene.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 310,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_001_seed_310_A polished metal ball gleaming under soft ambient.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_001_seed_42_A polished metal ball gleaming under soft ambient_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T03:04:04.527557",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_001_seed_42_A polished metal ball gleaming under soft ambient.mp4",
+  "arguments": {
+    "prompt": "A polished metal ball, gleaming under soft ambient lighting, begins its journey across a sleek, dark wooden tabletop. The camera remains fixed, capturing the ball's smooth, deliberate motion as it glides effortlessly, reflecting the room's subtle hues. The tabletop's surface, impeccably smooth, allows the ball to maintain a steady, unbroken path, creating a mesmerizing dance of light and shadow. As the ball rolls, its metallic surface mirrors the surroundings, offering fleeting glimpses of the room's decor. The gentle sound of the ball's movement adds a rhythmic element, enhancing the serene, captivating atmosphere of the scene.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 42,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_001_seed_42_A polished metal ball gleaming under soft ambient.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_001_seed_930_A polished metal ball gleaming under soft ambient_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-13T23:36:53.412174",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_001_seed_930_A polished metal ball gleaming under soft ambient.mp4",
+  "arguments": {
+    "prompt": "A polished metal ball, gleaming under soft ambient lighting, begins its journey across a sleek, dark wooden tabletop. The camera remains fixed, capturing the ball's smooth, deliberate motion as it glides effortlessly, reflecting the room's subtle hues. The tabletop's surface, impeccably smooth, allows the ball to maintain a steady, unbroken path, creating a mesmerizing dance of light and shadow. As the ball rolls, its metallic surface mirrors the surroundings, offering fleeting glimpses of the room's decor. The gentle sound of the ball's movement adds a rhythmic element, enhancing the serene, captivating atmosphere of the scene.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 930,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_001_seed_930_A polished metal ball gleaming under soft ambient.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_002_seed_310_In a sleek modern office with glass walls and mini_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T06:33:33.469770",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_002_seed_310_In a sleek modern office with glass walls and mini.mp4",
+  "arguments": {
+    "prompt": "In a sleek, modern office with glass walls and minimalist decor, two men stand facing each other, their expressions radiating warmth and camaraderie. One man, wearing a tailored navy suit with a crisp white shirt, extends his hand with a genuine smile. The other, dressed in a charcoal blazer and light blue shirt, reciprocates the gesture, their hands meeting in a firm, enthusiastic handshake. Sunlight streams through large windows, casting a soft glow on the polished wooden floor and contemporary furniture. The atmosphere is one of mutual respect and shared success, captured in this moment of professional connection.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 310,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_002_seed_310_In a sleek modern office with glass walls and mini.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_002_seed_42_In a sleek modern office with glass walls and mini_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T03:06:02.411960",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_002_seed_42_In a sleek modern office with glass walls and mini.mp4",
+  "arguments": {
+    "prompt": "In a sleek, modern office with glass walls and minimalist decor, two men stand facing each other, their expressions radiating warmth and camaraderie. One man, wearing a tailored navy suit with a crisp white shirt, extends his hand with a genuine smile. The other, dressed in a charcoal blazer and light blue shirt, reciprocates the gesture, their hands meeting in a firm, enthusiastic handshake. Sunlight streams through large windows, casting a soft glow on the polished wooden floor and contemporary furniture. The atmosphere is one of mutual respect and shared success, captured in this moment of professional connection.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 42,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_002_seed_42_In a sleek modern office with glass walls and mini.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_002_seed_930_In a sleek modern office with glass walls and mini_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-13T23:38:50.964717",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_002_seed_930_In a sleek modern office with glass walls and mini.mp4",
+  "arguments": {
+    "prompt": "In a sleek, modern office with glass walls and minimalist decor, two men stand facing each other, their expressions radiating warmth and camaraderie. One man, wearing a tailored navy suit with a crisp white shirt, extends his hand with a genuine smile. The other, dressed in a charcoal blazer and light blue shirt, reciprocates the gesture, their hands meeting in a firm, enthusiastic handshake. Sunlight streams through large windows, casting a soft glow on the polished wooden floor and contemporary furniture. The atmosphere is one of mutual respect and shared success, captured in this moment of professional connection.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 930,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_002_seed_930_In a sleek modern office with glass walls and mini.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_003_seed_310_A couple strolls hand in hand down a bustling city_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T06:35:31.033445",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_003_seed_310_A couple strolls hand in hand down a bustling city.mp4",
+  "arguments": {
+    "prompt": "A couple strolls hand in hand down a bustling city street, their fingers intertwined, embodying a sense of unity and affection. The man, wearing a navy coat and jeans, walks confidently beside the woman, who is dressed in a stylish red trench coat and black boots. The street is alive with the sounds of distant traffic and the chatter of passersby, while the couple remains in their own world, exchanging smiles and glances. The static camera captures their every step, as they navigate the urban landscape, framed by towering buildings and vibrant storefronts, under a sky painted with the warm hues of a setting sun.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 310,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_003_seed_310_A couple strolls hand in hand down a bustling city.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_003_seed_42_A couple strolls hand in hand down a bustling city_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T03:07:59.898208",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_003_seed_42_A couple strolls hand in hand down a bustling city.mp4",
+  "arguments": {
+    "prompt": "A couple strolls hand in hand down a bustling city street, their fingers intertwined, embodying a sense of unity and affection. The man, wearing a navy coat and jeans, walks confidently beside the woman, who is dressed in a stylish red trench coat and black boots. The street is alive with the sounds of distant traffic and the chatter of passersby, while the couple remains in their own world, exchanging smiles and glances. The static camera captures their every step, as they navigate the urban landscape, framed by towering buildings and vibrant storefronts, under a sky painted with the warm hues of a setting sun.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 42,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_003_seed_42_A couple strolls hand in hand down a bustling city.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_003_seed_930_A couple strolls hand in hand down a bustling city_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-13T23:40:49.450607",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_003_seed_930_A couple strolls hand in hand down a bustling city.mp4",
+  "arguments": {
+    "prompt": "A couple strolls hand in hand down a bustling city street, their fingers intertwined, embodying a sense of unity and affection. The man, wearing a navy coat and jeans, walks confidently beside the woman, who is dressed in a stylish red trench coat and black boots. The street is alive with the sounds of distant traffic and the chatter of passersby, while the couple remains in their own world, exchanging smiles and glances. The static camera captures their every step, as they navigate the urban landscape, framed by towering buildings and vibrant storefronts, under a sky painted with the warm hues of a setting sun.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 930,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_003_seed_930_A couple strolls hand in hand down a bustling city.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_004_seed_310_In a sunlit garden a young boy with tousled hair a_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T06:37:28.463026",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_004_seed_310_In a sunlit garden a young boy with tousled hair a.mp4",
+  "arguments": {
+    "prompt": "In a sunlit garden, a young boy with tousled hair and a bright smile leans forward, his eyes sparkling with joy as he extends his right hand. The camera captures the tender moment in close-up as his fingers gently stroke the soft, black-and-white fur of a Border Collie. The dog's eyes close in contentment, its tail wagging gently, reflecting the bond between them. The boy's laughter fills the air, harmonizing with the gentle rustle of leaves and distant birdsong, creating a serene and heartwarming scene of friendship and affection.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 310,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_004_seed_310_In a sunlit garden a young boy with tousled hair a.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_004_seed_42_In a sunlit garden a young boy with tousled hair a_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T03:09:57.461195",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_004_seed_42_In a sunlit garden a young boy with tousled hair a.mp4",
+  "arguments": {
+    "prompt": "In a sunlit garden, a young boy with tousled hair and a bright smile leans forward, his eyes sparkling with joy as he extends his right hand. The camera captures the tender moment in close-up as his fingers gently stroke the soft, black-and-white fur of a Border Collie. The dog's eyes close in contentment, its tail wagging gently, reflecting the bond between them. The boy's laughter fills the air, harmonizing with the gentle rustle of leaves and distant birdsong, creating a serene and heartwarming scene of friendship and affection.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 42,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_004_seed_42_In a sunlit garden a young boy with tousled hair a.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_004_seed_930_In a sunlit garden a young boy with tousled hair a_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-13T23:42:47.118049",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_004_seed_930_In a sunlit garden a young boy with tousled hair a.mp4",
+  "arguments": {
+    "prompt": "In a sunlit garden, a young boy with tousled hair and a bright smile leans forward, his eyes sparkling with joy as he extends his right hand. The camera captures the tender moment in close-up as his fingers gently stroke the soft, black-and-white fur of a Border Collie. The dog's eyes close in contentment, its tail wagging gently, reflecting the bond between them. The boy's laughter fills the air, harmonizing with the gentle rustle of leaves and distant birdsong, creating a serene and heartwarming scene of friendship and affection.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 930,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_004_seed_930_In a sunlit garden a young boy with tousled hair a.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_005_seed_310_The video begins with a rapid zoom towards a vibra_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T06:39:25.998868",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_005_seed_310_The video begins with a rapid zoom towards a vibra.mp4",
+  "arguments": {
+    "prompt": "The video begins with a rapid zoom towards a vibrant red rose, its petals glistening with morning dew under soft sunlight. As the camera closes in, the intricate details of the petals become visible, revealing their velvety texture and delicate, layered structure. The rich crimson hue of the petals contrasts beautifully with the subtle green of the surrounding leaves. The camera continues to zoom, capturing the gentle curves and natural patterns etched into each petal, highlighting the rose's elegance and complexity. The scene concludes with an intimate close-up of the rose's heart, showcasing the intricate beauty of nature.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 310,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_005_seed_310_The video begins with a rapid zoom towards a vibra.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_005_seed_42_The video begins with a rapid zoom towards a vibra_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T03:11:55.138217",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_005_seed_42_The video begins with a rapid zoom towards a vibra.mp4",
+  "arguments": {
+    "prompt": "The video begins with a rapid zoom towards a vibrant red rose, its petals glistening with morning dew under soft sunlight. As the camera closes in, the intricate details of the petals become visible, revealing their velvety texture and delicate, layered structure. The rich crimson hue of the petals contrasts beautifully with the subtle green of the surrounding leaves. The camera continues to zoom, capturing the gentle curves and natural patterns etched into each petal, highlighting the rose's elegance and complexity. The scene concludes with an intimate close-up of the rose's heart, showcasing the intricate beauty of nature.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 42,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_005_seed_42_The video begins with a rapid zoom towards a vibra.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_005_seed_930_The video begins with a rapid zoom towards a vibra_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-13T23:44:44.587390",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_005_seed_930_The video begins with a rapid zoom towards a vibra.mp4",
+  "arguments": {
+    "prompt": "The video begins with a rapid zoom towards a vibrant red rose, its petals glistening with morning dew under soft sunlight. As the camera closes in, the intricate details of the petals become visible, revealing their velvety texture and delicate, layered structure. The rich crimson hue of the petals contrasts beautifully with the subtle green of the surrounding leaves. The camera continues to zoom, capturing the gentle curves and natural patterns etched into each petal, highlighting the rose's elegance and complexity. The scene concludes with an intimate close-up of the rose's heart, showcasing the intricate beauty of nature.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 930,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_005_seed_930_The video begins with a rapid zoom towards a vibra.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_006_seed_310_An exhilarating FPV aerial shot captures the breat_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T06:41:23.367450",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_006_seed_310_An exhilarating FPV aerial shot captures the breat.mp4",
+  "arguments": {
+    "prompt": "An exhilarating FPV aerial shot captures the breathtaking panorama of snow-capped mountains, their majestic peaks glistening under the golden embrace of the morning sun. The camera swoops gracefully over the rugged terrain, revealing intricate patterns of shadows and light dancing across the pristine white snow. As the drone glides effortlessly through the crisp, clear air, the serene atmosphere envelops the scene, with the distant horizon painted in soft pastel hues. The tranquil silence of the mountains is palpable, broken only by the gentle whisper of the wind, creating a mesmerizing and peaceful visual symphony.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 310,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_006_seed_310_An exhilarating FPV aerial shot captures the breat.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_006_seed_42_An exhilarating FPV aerial shot captures the breat_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T03:13:52.538880",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_006_seed_42_An exhilarating FPV aerial shot captures the breat.mp4",
+  "arguments": {
+    "prompt": "An exhilarating FPV aerial shot captures the breathtaking panorama of snow-capped mountains, their majestic peaks glistening under the golden embrace of the morning sun. The camera swoops gracefully over the rugged terrain, revealing intricate patterns of shadows and light dancing across the pristine white snow. As the drone glides effortlessly through the crisp, clear air, the serene atmosphere envelops the scene, with the distant horizon painted in soft pastel hues. The tranquil silence of the mountains is palpable, broken only by the gentle whisper of the wind, creating a mesmerizing and peaceful visual symphony.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 42,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_006_seed_42_An exhilarating FPV aerial shot captures the breat.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_006_seed_930_An exhilarating FPV aerial shot captures the breat_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-13T23:46:41.940434",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_006_seed_930_An exhilarating FPV aerial shot captures the breat.mp4",
+  "arguments": {
+    "prompt": "An exhilarating FPV aerial shot captures the breathtaking panorama of snow-capped mountains, their majestic peaks glistening under the golden embrace of the morning sun. The camera swoops gracefully over the rugged terrain, revealing intricate patterns of shadows and light dancing across the pristine white snow. As the drone glides effortlessly through the crisp, clear air, the serene atmosphere envelops the scene, with the distant horizon painted in soft pastel hues. The tranquil silence of the mountains is palpable, broken only by the gentle whisper of the wind, creating a mesmerizing and peaceful visual symphony.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 930,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_006_seed_930_An exhilarating FPV aerial shot captures the breat.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_007_seed_310_A sleek black car glides smoothly along a winding_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T06:43:20.754622",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_007_seed_310_A sleek black car glides smoothly along a winding.mp4",
+  "arguments": {
+    "prompt": "A sleek black car glides smoothly along a winding coastal road, the camera capturing its every move with a handheld tracking shot. The sun casts a golden glow on the vehicle's polished surface, highlighting its aerodynamic design. As the car navigates sharp turns, the ocean waves crash against the rocky cliffs below, creating a dramatic backdrop. The camera follows closely, capturing the rhythmic motion of the wheels and the subtle reflections of the surrounding landscape on the car's glossy exterior. The scene conveys a sense of freedom and adventure, with the open road stretching endlessly ahead.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 310,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_007_seed_310_A sleek black car glides smoothly along a winding.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_007_seed_42_A sleek black car glides smoothly along a winding_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T03:15:49.970619",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_007_seed_42_A sleek black car glides smoothly along a winding.mp4",
+  "arguments": {
+    "prompt": "A sleek black car glides smoothly along a winding coastal road, the camera capturing its every move with a handheld tracking shot. The sun casts a golden glow on the vehicle's polished surface, highlighting its aerodynamic design. As the car navigates sharp turns, the ocean waves crash against the rocky cliffs below, creating a dramatic backdrop. The camera follows closely, capturing the rhythmic motion of the wheels and the subtle reflections of the surrounding landscape on the car's glossy exterior. The scene conveys a sense of freedom and adventure, with the open road stretching endlessly ahead.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 42,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_007_seed_42_A sleek black car glides smoothly along a winding.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_007_seed_930_A sleek black car glides smoothly along a winding_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-13T23:48:39.387827",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_007_seed_930_A sleek black car glides smoothly along a winding.mp4",
+  "arguments": {
+    "prompt": "A sleek black car glides smoothly along a winding coastal road, the camera capturing its every move with a handheld tracking shot. The sun casts a golden glow on the vehicle's polished surface, highlighting its aerodynamic design. As the car navigates sharp turns, the ocean waves crash against the rocky cliffs below, creating a dramatic backdrop. The camera follows closely, capturing the rhythmic motion of the wheels and the subtle reflections of the surrounding landscape on the car's glossy exterior. The scene conveys a sense of freedom and adventure, with the open road stretching endlessly ahead.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 930,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_007_seed_930_A sleek black car glides smoothly along a winding.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_008_seed_310_From a high vantage point the camera captures a bu_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T06:45:18.349837",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_008_seed_310_From a high vantage point the camera captures a bu.mp4",
+  "arguments": {
+    "prompt": "From a high vantage point, the camera captures a bustling city street below, where pedestrians in colorful attire navigate the crosswalks, their movements synchronized like a choreographed dance. The soft, golden lighting of the late afternoon bathes the scene, casting long shadows and highlighting the vibrant hues of the cityscape. Cars glide smoothly along the road, their headlights twinkling like stars in the gentle light. The camera pans slowly, revealing a tapestry of urban life, where the rhythm of footsteps and the hum of traffic create a harmonious symphony, all under the warm embrace of the setting sun.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 310,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_008_seed_310_From a high vantage point the camera captures a bu.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_008_seed_42_From a high vantage point the camera captures a bu_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T03:17:47.396086",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_008_seed_42_From a high vantage point the camera captures a bu.mp4",
+  "arguments": {
+    "prompt": "From a high vantage point, the camera captures a bustling city street below, where pedestrians in colorful attire navigate the crosswalks, their movements synchronized like a choreographed dance. The soft, golden lighting of the late afternoon bathes the scene, casting long shadows and highlighting the vibrant hues of the cityscape. Cars glide smoothly along the road, their headlights twinkling like stars in the gentle light. The camera pans slowly, revealing a tapestry of urban life, where the rhythm of footsteps and the hum of traffic create a harmonious symphony, all under the warm embrace of the setting sun.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 42,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_008_seed_42_From a high vantage point the camera captures a bu.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_008_seed_930_From a high vantage point the camera captures a bu_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-13T23:50:36.750241",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_008_seed_930_From a high vantage point the camera captures a bu.mp4",
+  "arguments": {
+    "prompt": "From a high vantage point, the camera captures a bustling city street below, where pedestrians in colorful attire navigate the crosswalks, their movements synchronized like a choreographed dance. The soft, golden lighting of the late afternoon bathes the scene, casting long shadows and highlighting the vibrant hues of the cityscape. Cars glide smoothly along the road, their headlights twinkling like stars in the gentle light. The camera pans slowly, revealing a tapestry of urban life, where the rhythm of footsteps and the hum of traffic create a harmonious symphony, all under the warm embrace of the setting sun.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 930,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_008_seed_930_From a high vantage point the camera captures a bu.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_009_seed_310_From a high vantage point the bustling street belo_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T06:47:15.988138",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_009_seed_310_From a high vantage point the bustling street belo.mp4",
+  "arguments": {
+    "prompt": "From a high vantage point, the bustling street below is alive with activity, as pedestrians in colorful attire navigate the wide, sunlit avenue. The bright lighting casts sharp shadows, highlighting the diversity of people, from a woman in a red coat carrying shopping bags to a man in a business suit briskly walking. The scene captures the essence of urban life, with a cyclist weaving through the crowd and a street performer playing a lively tune on a saxophone. The vibrant energy is palpable, as the golden sunlight bathes the scene, creating a dynamic interplay of light and movement.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 310,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_009_seed_310_From a high vantage point the bustling street belo.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_009_seed_42_From a high vantage point the bustling street belo_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T03:19:44.892415",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_009_seed_42_From a high vantage point the bustling street belo.mp4",
+  "arguments": {
+    "prompt": "From a high vantage point, the bustling street below is alive with activity, as pedestrians in colorful attire navigate the wide, sunlit avenue. The bright lighting casts sharp shadows, highlighting the diversity of people, from a woman in a red coat carrying shopping bags to a man in a business suit briskly walking. The scene captures the essence of urban life, with a cyclist weaving through the crowd and a street performer playing a lively tune on a saxophone. The vibrant energy is palpable, as the golden sunlight bathes the scene, creating a dynamic interplay of light and movement.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 42,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_009_seed_42_From a high vantage point the bustling street belo.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_009_seed_930_From a high vantage point the bustling street belo_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-13T23:52:34.173912",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_009_seed_930_From a high vantage point the bustling street belo.mp4",
+  "arguments": {
+    "prompt": "From a high vantage point, the bustling street below is alive with activity, as pedestrians in colorful attire navigate the wide, sunlit avenue. The bright lighting casts sharp shadows, highlighting the diversity of people, from a woman in a red coat carrying shopping bags to a man in a business suit briskly walking. The scene captures the essence of urban life, with a cyclist weaving through the crowd and a street performer playing a lively tune on a saxophone. The vibrant energy is palpable, as the golden sunlight bathes the scene, creating a dynamic interplay of light and movement.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 930,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_009_seed_930_From a high vantage point the bustling street belo.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_010_seed_310_From a high vantage point the camera captures a bu_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T06:49:13.551769",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_010_seed_310_From a high vantage point the camera captures a bu.mp4",
+  "arguments": {
+    "prompt": "From a high vantage point, the camera captures a bustling urban street scene at dusk, where pedestrians, clad in winter coats and scarves, navigate the wet pavement, their reflections shimmering in the dim, golden streetlights. The soft glow from nearby shop windows casts a warm hue, contrasting with the cool, misty air. Cars, with headlights on, move slowly, their tires splashing through shallow puddles. The atmosphere is serene yet lively, with the distant hum of city life and the occasional sound of footsteps echoing through the crisp evening air, creating a harmonious urban symphony.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 310,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val2/step_000100/sample_010_seed_310_From a high vantage point the camera captures a bu.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}

outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val/step_000100/sample_010_seed_42_From a high vantage point the camera captures a bu_config.json

@@ -0,0 +1,49 @@
+{
+  "timestamp": "2026-05-14T03:21:42.384011",
+  "task": "t2v",
+  "transformer_version": "480p_t2v",
+  "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_010_seed_42_From a high vantage point the camera captures a bu.mp4",
+  "arguments": {
+    "prompt": "From a high vantage point, the camera captures a bustling urban street scene at dusk, where pedestrians, clad in winter coats and scarves, navigate the wet pavement, their reflections shimmering in the dim, golden streetlights. The soft glow from nearby shop windows casts a warm hue, contrasting with the cool, misty air. Cars, with headlights on, move slowly, their tires splashing through shallow puddles. The atmosphere is serene yet lively, with the distant hum of city life and the occasional sound of footsteps echoing through the crisp evening air, creating a harmonious urban symphony.",
+    "valid_video_csv": [
+      "assets/video_gen_eval_top100.csv"
+    ],
+    "negative_prompt": "",
+    "resolution": "480p",
+    "model_path": "./ckpts",
+    "aspect_ratio": "16:9",
+    "fixed_size": "480x864",
+    "num_inference_steps": 40,
+    "video_length": 121,
+    "sr": false,
+    "save_pre_sr_video": false,
+    "rewrite": false,
+    "cfg_distilled": false,
+    "enable_step_distill": false,
+    "sparse_attn": false,
+    "offloading": false,
+    "group_offloading": null,
+    "overlap_group_offloading": false,
+    "dtype": "bf16",
+    "seed": 42,
+    "image_path": null,
+    "output_path": "./outputs/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/videogen_val1/step_000100/sample_010_seed_42_From a high vantage point the camera captures a bu.mp4",
+    "original_noise_dir": null,
+    "use_sageattn": false,
+    "sage_blocks_range": "0-53",
+    "enable_torch_compile": false,
+    "enable_cache": false,
+    "cache_type": "deepcache",
+    "no_cache_block_id": "53",
+    "cache_start_step": 11,
+    "cache_end_step": 45,
+    "total_steps": 50,
+    "cache_step_interval": 4,
+    "save_generation_config": true,
+    "checkpoint_path": "/root/yw006-4-sage-qwen35-9b-token-credit-video-adv-2/checkpoint-100/transformer",
+    "lora_path": null,
+    "use_fp8_gemm": false,
+    "quant_type": "fp8-per-token-sgl",
+    "include_patterns": "double_blocks"
+  }
+}