Video Understanding on Video-MME w/ Subtitles
61.33ScoreLLaVA-Video-7B
Evaluation results
| Method | Links | ||
|---|---|---|---|
| LLaVA-Video-7BVisual Token Budget=All 2,704 Visual Tokens, Evaluation Mode=16-frame evaluation, Base Model=LLaVA-Video-7B2026.07 | 61.33 | 100 | |
| AnchorPruneVisual Token Budget=Retain 1,024 Tokens, Evaluation Mode=16-frame evaluation, Base Model=LLaVA-Video-7B2026.07 | 60.89 | 98.3 | |
| FastVVisual Token Budget=Retain 1,024 Tokens, Evaluation Mode=16-frame evaluation, Base Model=LLaVA-Video-7B, Source=ECCV’242026.07 | 60.59 | 97.7 | |
| CDPrunerVisual Token Budget=Retain 1,024 Tokens, Evaluation Mode=16-frame evaluation, Base Model=LLaVA-Video-7B, Source=NeurIPS’252026.07 | 59.81 | 96.8 | |
| DivPruneVisual Token Budget=Retain 1,024 Tokens, Evaluation Mode=16-frame evaluation, Base Model=LLaVA-Video-7B, Source=CVPR’252026.07 | 59.15 | 96.1 | |
| DivPruneVisual Token Budget=Retain 512 Tokens, Evaluation Mode=16-frame evaluation, Base Model=LLaVA-Video-7B, Source=CVPR’252026.07 | 58.19 | 93.4 | |
| AnchorPruneVisual Token Budget=Retain 512 Tokens, Evaluation Mode=16-frame evaluation, Base Model=LLaVA-Video-7B2026.07 | 58.07 | 94.1 | |
| FastVVisual Token Budget=Retain 512 Tokens, Evaluation Mode=16-frame evaluation, Base Model=LLaVA-Video-7B, Source=ECCV’242026.07 | 57.63 | 92.4 | |
| CDPrunerVisual Token Budget=Retain 512 Tokens, Evaluation Mode=16-frame evaluation, Base Model=LLaVA-Video-7B, Source=NeurIPS’252026.07 | 57 | 91.7 |