Long Video Understanding on MLVU m-avg
69.8MLVU m-avg ScoreReflect-R1
Evaluation Results
| Method | Links | |
|---|---|---|
| Reflect-R1# Frame=768, Paradigm=Tool-Augmented Reasoning, Adaptive keyframe retrieval=false2026.06 | 69.8 | |
| TimeSearch-R-7B# Frame=768, Paradigm=Tool-Augmented Reasoning, Adaptive keyframe retrieval=false2026.06 | 69.3 | |
| Video-R1-7B# Frame=768, Paradigm=Standard (w/o Tools), Adaptive keyframe retrieval=false2026.06 | 68.5 | |
| Vision-R1-7B# Frame=768, Paradigm=Standard (w/o Tools), Adaptive keyframe retrieval=false2026.06 | 68.2 | |
| GPT-4o# Frame=384, Paradigm=Standard (w/o Tools), Adaptive keyframe retrieval=false2026.06 | 64.6 | |
| VL-Rethinker# Frame=768, Paradigm=Standard (w/o Tools), Adaptive keyframe retrieval=false2026.06 | 63.5 | |
| Qwen2.5-VL-7B-Instruct# Frame=768, Paradigm=Standard (w/o Tools), Adaptive keyframe retrieval=false2026.06 | 57.9 |