Localization on RefCOCOg
70.7AccuracyVanilla
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| VanillaVisual Token Budget=576, FLOPs (T)=4.292026.02 | 70.7 | 100 | |
| SwiftVLMVisual Token Budget=192, FLOPs (T)=1.752026.02 | 60.6 | 85.7 | |
| SwiftVLMVisual Token Budget=128, FLOPs (T)=1.312026.02 | 47.4 | 67 | |
| FEATHERVisual Token Budget=192, FLOPs (T)=1.822026.02 | 45.4 | 64.2 | |
| FEATHERVisual Token Budget=128, FLOPs (T)=1.442026.02 | 35.2 | 49.8 | |
| FastVVisual Token Budget=192, FLOPs (T)=1.712026.02 | 29.7 | 42 | |
| PDropVisual Token Budget=192, FLOPs (T)=1.722026.02 | 18.7 | 26.4 | |
| FastVVisual Token Budget=128, FLOPs (T)=1.292026.02 | 13.8 | 19.5 | |
| SparseVLMVisual Token Budget=192, FLOPs (T)=1.722026.02 | 7.1 | 10 | |
| VisionZipVisual Token Budget=192, FLOPs (T)=1.712026.02 | 6.3 | 8.9 | |
| VisionZipVisual Token Budget=128, FLOPs (T)=1.292026.02 | 4.3 | 6.1 | |
| SparseVLMVisual Token Budget=128, FLOPs (T)=1.32026.02 | 4.1 | 5.8 | |
| PDropVisual Token Budget=128, FLOPs (T)=1.282026.02 | 2.3 | 3.3 |