Visual Question Answering on TextVQA (Accuracy, Average)
60.3AccuracyLLaVA-NeXT-7B
Evaluation results
| Method | Links | ||
|---|---|---|---|
| LLaVA-NeXT-7BBase Model=LLaVA-NeXT-7B, Retained Tokens=28802026.03 | 60.3 | 100 | |
| VisionZipBase Model=LLaVA-NeXT-7B, Retained Tokens=6402026.03 | 59.9 | 99.8 | |
| SparseVLMBase Model=LLaVA-NeXT-7B, Retained Tokens=6402026.03 | 59.7 | 98.8 | |
| AgilePrunerBase Model=LLaVA-NeXT-7B, Retained Tokens=6402026.03 | 59 | 99.64 | |
| VisionZipBase Model=LLaVA-NeXT-7B, Retained Tokens=3202026.03 | 58.8 | 96.4 | |
| AgilePrunerBase Model=LLaVA-NeXT-7B, Retained Tokens=3202026.03 | 58.4 | 97.94 | |
| FastVBase Model=LLaVA-NeXT-7B, Retained Tokens=6402026.03 | 58.1 | 95.2 | |
| PDropBase Model=LLaVA-NeXT-7B, Retained Tokens=6402026.03 | 57.8 | 97 | |
| DivPruneBase Model=LLaVA-NeXT-7B, Retained Tokens=6402026.03 | 57 | 98.9 | |
| SparseVLMBase Model=LLaVA-NeXT-7B, Retained Tokens=3202026.03 | 56.5 | 94.6 | |
| DivPruneBase Model=LLaVA-NeXT-7B, Retained Tokens=3202026.03 | 56.2 | 97 | |
| PruMerge+Base Model=LLaVA-NeXT-7B, Retained Tokens=6402026.03 | 54.9 | 98.3 | |
| PruMerge+Base Model=LLaVA-NeXT-7B, Retained Tokens=3202026.03 | 54 | 95.7 | |
| FastVBase Model=LLaVA-NeXT-7B, Retained Tokens=3202026.03 | 52.2 | 79.9 | |
| PDropBase Model=LLaVA-NeXT-7B, Retained Tokens=3202026.03 | 49 | 82.5 |