Closed-ended Visual Question Answering on PitVQA
60.43F1 ScoreZEN
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| ZENBackbone=ViT-B, Protocol=Fine-tuning backbone2026.02 | 60.43 | 60.71 | 61.9 | |
| MIS-DINOv2Backbone=ViT-L, Protocol=Fine-tuning backbone2026.02 | 59.2 | 59.28 | 60.38 | |
| SurgeNetBackbone=CAFormer18, Protocol=Fine-tuning backbone2026.02 | 59.17 | 59.84 | 60.94 | |
| MIS-DINOv2Backbone=ViT-B, Protocol=Fine-tuning backbone2026.02 | 56.58 | 56.68 | 57.85 | |
| MIS-DINOv2Backbone=ViT-L, Protocol=Frozen backbone2026.02 | 56.1 | 56.8 | 58.51 | |
| MIS-DINOBackbone=ViT-B, Protocol=Fine-tuning backbone2026.02 | 55.23 | 55.68 | 57.15 | |
| ImageNetBackbone=ViT-B, Protocol=Fine-tuning backbone2026.02 | 55.23 | 56.17 | 57.49 | |
| ZENBackbone=ViT-B, Protocol=Frozen backbone2026.02 | 55.1 | 55.37 | 56.73 | |
| MIS-MAEBackbone=ViT-B, Protocol=Fine-tuning backbone2026.02 | 54.65 | 55.49 | 56.84 | |
| MIS-DINOv2Backbone=ViT-B, Protocol=Frozen backbone2026.02 | 54.47 | 54.46 | 55.85 | |
| MIS-MoCoV3Backbone=ViT-B, Protocol=Fine-tuning backbone2026.02 | 53.81 | 54.25 | 55.63 | |
| PeskaVLPBackbone=ResNet50, Protocol=Fine-tuning backbone2026.02 | 53.64 | 54.52 | 55.9 | |
| MIS-DINOBackbone=ViT-B, Protocol=Frozen backbone2026.02 | 52.81 | 52.9 | 54.47 | |
| EndoFMBackbone=ViT-B, Protocol=Fine-tuning backbone2026.02 | 52.09 | 53.06 | 54.64 | |
| EndoViTBackbone=ViT-B, Protocol=Fine-tuning backbone2026.02 | 50.78 | 52.21 | 54.13 | |
| MIS-MoCoV3Backbone=ViT-B, Protocol=Frozen backbone2026.02 | 47.83 | 49.23 | 51.27 | |
| ImageNetBackbone=ViT-B, Protocol=Frozen backbone2026.02 | 47.7 | 49.52 | 51.21 | |
| EndoFMBackbone=ViT-B, Protocol=Frozen backbone2026.02 | 47.59 | 48.2 | 49.95 | |
| SurgeNetBackbone=CAFormer18, Protocol=Frozen backbone2026.02 | 47.08 | 48.46 | 50.18 | |
| PeskaVLPBackbone=ResNet50, Protocol=Frozen backbone2026.02 | 44.38 | 45.55 | 47.38 | |
| MIS-MAEBackbone=ViT-B, Protocol=Frozen backbone2026.02 | 42.91 | 46.19 | 48.52 | |
| EndoViTBackbone=ViT-B, Protocol=Frozen backbone2026.02 | 41.63 | 44.26 | 46.5 | |
| MIS-MSNBackbone=ViT-B, Protocol=Fine-tuning backbone2026.02 | 40.47 | 42.21 | 43.57 | |
| MIS-MSNBackbone=ViT-B, Protocol=Frozen backbone2026.02 | 27.08 | 28.48 | 29.43 | |
| GSViTBackbone=EfficientViT, Protocol=Frozen backbone2026.02 | 25.74 | 27.37 | 29.57 | |
| GSViTBackbone=EfficientViT, Protocol=Fine-tuning backbone2026.02 | 24.9 | 26.81 | 29.03 |