Video Understanding on VideoMME no subtitles (test)
71.9AccuracyGPT-4o
Evaluation results
| Method | Links | |
|---|---|---|
| GPT-4oFrames=-2026.04 | 71.9 | |
| Video-ToCFrames=642026.04 | 62.6 | |
| Video-R1Frames=642026.04 | 61.4 | |
| Video-ToCFrames=322026.04 | 61.2 | |
| Video-ToC-SFTFrames=642026.04 | 61.1 | |
| Video-ToC-SFTFrames=322026.04 | 59.8 | |
| Baseline (Qwen2.5-VL-7B)Frames=642026.04 | 59.6 | |
| Video-R1Frames=322026.04 | 59.3 | |
| Video-R1-SFTFrames=642026.04 | 58.8 | |
| Video-ToCFrames=162026.04 | 58.6 | |
| LLaVA-OneVision-7BFrames=-2026.04 | 58.2 | |
| VideoChat-R1Frames=162026.04 | 57.7 | |
| Video-R1Frames=162026.04 | 57.4 | |
| Video-ToC-SFTFrames=162026.04 | 56.6 | |
| Baseline (Qwen2.5-VL-7B)Frames=322026.04 | 56.6 | |
| Video-R1-SFTFrames=322026.04 | 55.4 | |
| Baseline (Qwen2.5-VL-7B)Frames=162026.04 | 53.1 | |
| Video-R1-SFTFrames=162026.04 | 52.8 | |
| LongVA-7BFrames=-2026.04 | 52.6 | |
| VideoLLaMA2-7BFrames=-2026.04 | 47.9 | |
| TinyLLaVA-Video-R1Frames=162026.04 | 46.6 |