Visual Language Model Evaluation on MMVet
40.6MMVet ScoreDART
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DARTModel=LLaVA-Next-7B, Token Budget=320, Reduction Ratio=88.9%2026.05 | 40.6 | 96.4 | |
| VanillaModel=LLaVA-Next-7B, Token Budget=2880, Reduction Ratio=0%2026.05 | 40 | 100 | |
| DivPruneModel=LLaVA-Next-7B, Token Budget=320, Reduction Ratio=88.9%2026.05 | 38.9 | 96.1 | |
| VisionZipModel=LLaVA-Next-7B, Token Budget=320, Reduction Ratio=88.9%2026.05 | 38.8 | 96.2 | |
| SPprunerModel=LLaVA-Next-7B, Token Budget=320, Reduction Ratio=88.9%2026.05 | 38.1 | 97.7 | |
| PyramidDropModel=LLaVA-Next-7B, Token Budget=320, Reduction Ratio=88.9%2026.05 | 30.2 | 90.1 | |
| FastVModel=LLaVA-Next-7B, Token Budget=320, Reduction Ratio=88.9%2026.05 | 25 | 86.8 |