Vision-Language Perception on MME Perception
1,389.5Perception ScoreSigLIP2-B
Evaluation results
| Method | Links | |
|---|---|---|
| SigLIP2-BBackbone=ViT-B2026.03 | 1,389.5 | |
| PEcore-BBackbone=ViT-B2026.03 | 1,375.5 | |
| EUPE-ViT-BBackbone=ViT-B2026.03 | 1,374.5 | |
| DINOv3-ViT-BBackbone=ViT-B2026.03 | 1,368 | |
| RADIOv2.5-BBackbone=ViT-B2026.03 | 1,349.8 | |
| DUNE-BBackbone=ViT-B2026.03 | 1,294.8 | |
| PEspatial-BBackbone=ViT-B2026.03 | 1,279.6 |