Closed-ended Visual Question Answering on LLS48-VQA
23.35F1 ScoreZEN
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| ZENBackbone=ViT-B, Evaluation Protocol=Fine-tuning backbone2026.02 | 23.35 | 28.06 | 40.06 | |
| MIS-DINOv2Backbone=ViT-L, Evaluation Protocol=Fine-tuning backbone2026.02 | 21.74 | 26.64 | 40.2 | |
| MIS-DINOv2Backbone=ViT-L, Evaluation Protocol=Frozen backbone2026.02 | 21.3 | 26 | 40.27 | |
| ZENBackbone=ViT-B, Evaluation Protocol=Frozen backbone2026.02 | 20.76 | 25.32 | 39.5 | |
| SurgeNetBackbone=CAFormer18, Evaluation Protocol=Fine-tuning backbone2026.02 | 20.48 | 23.96 | 34.25 | |
| MIS-DINOv2Backbone=ViT-B, Evaluation Protocol=Fine-tuning backbone2026.02 | 20.41 | 24.59 | 37.91 | |
| MIS-DINOBackbone=ViT-B, Evaluation Protocol=Fine-tuning backbone2026.02 | 19.44 | 23.05 | 34.84 | |
| MIS-MAEBackbone=ViT-B, Evaluation Protocol=Fine-tuning backbone2026.02 | 19.34 | 22.86 | 34.47 | |
| MIS-DINOv2Backbone=ViT-B, Evaluation Protocol=Frozen backbone2026.02 | 19.23 | 23.22 | 36.99 | |
| MIS-DINOBackbone=ViT-B, Evaluation Protocol=Frozen backbone2026.02 | 18.83 | 22.61 | 36.4 | |
| PeskaVLPBackbone=ResNet50, Evaluation Protocol=Fine-tuning backbone2026.02 | 17.64 | 20.85 | 32.73 | |
| MIS-MoCoV3Backbone=ViT-B, Evaluation Protocol=Frozen backbone2026.02 | 17.3 | 19.44 | 25.84 | |
| MIS-MoCoV3Backbone=ViT-B, Evaluation Protocol=Fine-tuning backbone2026.02 | 17.18 | 19.81 | 27.01 | |
| SurgeNetBackbone=CAFormer18, Evaluation Protocol=Frozen backbone2026.02 | 16.46 | 20.13 | 33.58 | |
| EndoViTBackbone=ViT-B, Evaluation Protocol=Fine-tuning backbone2026.02 | 16.45 | 19.04 | 29.51 | |
| ImageNetBackbone=ViT-B, Evaluation Protocol=Fine-tuning backbone2026.02 | 15.95 | 18.98 | 31.39 | |
| PeskaVLPBackbone=ResNet50, Evaluation Protocol=Frozen backbone2026.02 | 15.28 | 17.96 | 30.02 | |
| ImageNetBackbone=ViT-B, Evaluation Protocol=Frozen backbone2026.02 | 14.83 | 17.8 | 31.6 | |
| MIS-MAEBackbone=ViT-B, Evaluation Protocol=Frozen backbone2026.02 | 14.57 | 17.42 | 29.57 | |
| EndoFMBackbone=ViT-B, Evaluation Protocol=Fine-tuning backbone2026.02 | 14.31 | 17.25 | 30.98 | |
| EndoFMBackbone=ViT-B, Evaluation Protocol=Frozen backbone2026.02 | 13.75 | 16.8 | 31.3 | |
| EndoViTBackbone=ViT-B, Evaluation Protocol=Frozen backbone2026.02 | 13.66 | 15.36 | 25.39 | |
| MIS-MSNBackbone=ViT-B, Evaluation Protocol=Fine-tuning backbone2026.02 | 12.4 | 14.42 | 15.7 | |
| GSViTBackbone=EfficientViT, Evaluation Protocol=Fine-tuning backbone2026.02 | 5.68 | 7.34 | 7.8 | |
| GSViTBackbone=EfficientViT, Evaluation Protocol=Frozen backbone2026.02 | 5.26 | 6.83 | 7.38 | |
| MIS-MSNBackbone=ViT-B, Evaluation Protocol=Frozen backbone2026.02 | 3.99 | 6.06 | 6.06 |