Visual Question Answering on MMBench (Accuracy, Avg. Ratio)
64.7AccuracyVanilla
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| VanillaSource=CVPR’24, Average Token=576, Reroute=false2026.06 | 64.7 | 100 | |
| NüwaSource=ICLR’26, Average Token=192, Reroute=false2026.06 | 64.3 | 99.2 | |
| Nüwa+RerouteSource=Ours, Average Token=192, Reroute=true2026.06 | 63.6 | 100.6 | |
| NüwaSource=ICLR’26, Average Token=128, Reroute=false2026.06 | 63.4 | 98.4 | |
| PDropSource=CVPR’25, Average Token=192, Reroute=false2026.06 | 63.2 | 95.7 | |
| Nüwa+RerouteSource=Ours, Average Token=128, Reroute=true2026.06 | 63.2 | 99.8 | |
| VisionZipSource=CVPR’25, Average Token=192, Reroute=false2026.06 | 63 | 97.9 | |
| SparseVLMSource=ICML’25, Average Token=192, Reroute=false2026.06 | 62.5 | 95.4 | |
| VisionZipSource=CVPR’25, Average Token=128, Reroute=false2026.06 | 62 | 96.9 | |
| NüwaSource=ICLR’26, Average Token=64, Reroute=false2026.06 | 62 | 95.8 | |
| Nüwa+RerouteSource=Ours, Average Token=64, Reroute=true2026.06 | 61.9 | 96.7 | |
| FastVSource=ECCV’24, Average Token=192, Reroute=false2026.06 | 61.2 | 90.1 | |
| PDropSource=CVPR’25, Average Token=128, Reroute=false2026.06 | 61.1 | 93.8 | |
| SparseVLMSource=ICML’25, Average Token=64, Reroute=false2026.06 | 60.1 | 91.5 | |
| VisionZipSource=CVPR’25, Average Token=64, Reroute=false2026.06 | 60.1 | 93.2 | |
| SparseVLMSource=ICML’25, Average Token=128, Reroute=false2026.06 | 60 | 92.9 | |
| PruMergeSource=ICCV’25, Average Token=128, Reroute=false2026.06 | 59.6 | 94.7 | |
| PruMergeSource=ICCV’25, Average Token=64, Reroute=false2026.06 | 59.6 | 92.1 | |
| FastVSource=ECCV’24, Average Token=128, Reroute=false2026.06 | 56.1 | 85.2 | |
| FastVSource=ECCV’24, Average Token=64, Reroute=false2026.06 | 48 | 77.9 | |
| PDropSource=CVPR’25, Average Token=64, Reroute=false2026.06 | 33.3 | 70.3 |