Vision-Language Perception and Cognition on MME
1,851Total ScoreVanilla
Evaluation results
| Method | Links | ||
|---|---|---|---|
| VanillaBackbone=LLaVA-NeXT-7B, Token Budget=2880 Tokens, Venue=-2026.02 | 1,851 | — | |
| PIO-FVLMBackbone=LLaVA-NeXT-7B, Token Budget=320 Tokens, Venue=Ours2026.02 | 1,795 | — | |
| HoloVBackbone=LLaVA-NeXT-7B, Token Budget=320 Tokens, Venue=NeurIPS’252026.02 | 1,738 | — | |
| DARTBackbone=LLaVA-NeXT-7B, Token Budget=320 Tokens, Venue=EMNLP’252026.02 | 1,710 | — | |
| VisionZipBackbone=LLaVA-NeXT-7B, Token Budget=320 Tokens, Venue=CVPR’252026.02 | 1,702 | — | |
| FastVBackbone=LLaVA-NeXT-7B, Token Budget=320 Tokens, Venue=ECCV’242026.02 | 1,661 | — | |
| SparseVLMBackbone=LLaVA-NeXT-7B, Token Budget=320 Tokens, Venue=ICML’252026.02 | 1,533 | — | |
| CDPrunerBackbone=LLaVA-NeXT-7B, Token Budget=320 Tokens, Venue=NeurIPS’252026.02 | 1,453 | — |