Multi-skill Compositional Reasoning on MM-Vet
32.8AccuracySPARE
Evaluation results
| Method | Links | |
|---|---|---|
| SPAREBackbone=LLaVA-1.5-7B, Retained Tokens=1282026.06 | 32.8 | |
| CDPrunerBackbone=LLaVA-1.5-7B, Retained Tokens=1282026.06 | 32 | |
| SPAREBackbone=LLaVA-1.5-7B, Retained Tokens=642026.06 | 31.6 | |
| SPAREBackbone=LLaVA-1.5-7B, Retained Tokens=322026.06 | 31.4 | |
| LLaVA-1.5-7BBackbone=LLaVA-1.5-7B, Retained Tokens=5762026.06 | 31.3 | |
| CDPrunerBackbone=LLaVA-1.5-7B, Retained Tokens=642026.06 | 29.8 | |
| SparseVLMBackbone=LLaVA-1.5-7B, Retained Tokens=1282026.06 | 28.9 | |
| DARTBackbone=LLaVA-1.5-7B, Retained Tokens=1282026.06 | 28.7 | |
| CDPrunerBackbone=LLaVA-1.5-7B, Retained Tokens=322026.06 | 27.9 | |
| PruMerge+Backbone=LLaVA-1.5-7B, Retained Tokens=1282026.06 | 27.7 | |
| DARTBackbone=LLaVA-1.5-7B, Retained Tokens=642026.06 | 26.6 | |
| PruMerge+Backbone=LLaVA-1.5-7B, Retained Tokens=642026.06 | 26.3 | |
| SparseVLMBackbone=LLaVA-1.5-7B, Retained Tokens=642026.06 | 24.3 | |
| DARTBackbone=LLaVA-1.5-7B, Retained Tokens=322026.06 | 22.7 |