Visual Question Answering on MMVet (test)
67.1ScoreVanilla
Evaluation results
| Method | Links | |
|---|---|---|
| VanillaMode=Upper bound2024.12 | 67.1 | |
| iLLaVAImage token reduction ratio=66.7%2024.12 | 66 | |
| iLLaVAImage token reduction ratio=77.8%2024.12 | 65.2 | |
| VisionZipImage token reduction ratio=66.7%2024.12 | 64.8 | |
| SparseVLMImage token reduction ratio=66.7%2024.12 | 64.6 | |
| PyramidDropImage token reduction ratio=66.7%2024.12 | 64.3 | |
| FasterVLMImage token reduction ratio=66.7%2024.12 | 63.9 | |
| FasterVLMImage token reduction ratio=77.8%2024.12 | 63.7 | |
| VisionZipImage token reduction ratio=77.8%2024.12 | 63.1 | |
| iLLaVAImage token reduction ratio=88.9%2024.12 | 63 | |
| SparseVLMImage token reduction ratio=77.8%2024.12 | 62.9 | |
| FasterVLMImage token reduction ratio=88.9%2024.12 | 62.2 | |
| VisionZipImage token reduction ratio=88.9%2024.12 | 62 | |
| PyramidDropImage token reduction ratio=77.8%2024.12 | 61.7 | |
| SparseVLMImage token reduction ratio=88.9%2024.12 | 60.8 | |
| PyramidDropImage token reduction ratio=88.9%2024.12 | 57.3 |