Medical Image Segmentation on Site B
81.4Dice ScoreMCDRL
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| MCDRLBackbone=ViT-L/14, Pre-trained weights=CLIP2025.08 | 81.4 | 73.9 | 93.8 | |
| MCDRLBackbone=ViT-B/16, Pre-trained weights=CLIP2025.08 | 79.9 | 72.6 | 93 | |
| BiomedCoOpBackbone=ViT-L/14, Pre-trained weights=CLIP2025.08 | 79.3 | 72.1 | 92.7 | |
| MCDRLBackbone=ResNet-50, Pre-trained weights=CLIP2025.08 | 78.5 | 71.3 | 92.1 | |
| StyLIPBackbone=ViT-L/14, Pre-trained weights=CLIP2025.08 | 78.2 | 70.9 | 92.1 | |
| BiomedCoOpBackbone=ViT-B/16, Pre-trained weights=CLIP2025.08 | 77.9 | 70.8 | 91.9 | |
| StyLIPBackbone=ViT-B/16, Pre-trained weights=CLIP2025.08 | 76.8 | 69.7 | 91.3 | |
| BiomedCoOpBackbone=ResNet-50, Pre-trained weights=CLIP2025.08 | 76.6 | 69.7 | 90.8 | |
| StyLIPBackbone=ResNet-50, Pre-trained weights=CLIP2025.08 | 75.4 | 68.2 | 90.2 | |
| BaselineBackbone=ViT-L/14, Pre-trained weights=CLIP2025.08 | 75.2 | 68.2 | 90.4 | |
| BaselineBackbone=ViT-B/16, Pre-trained weights=CLIP2025.08 | 73.5 | 66.7 | 89.5 | |
| BaselineBackbone=ResNet-50, Pre-trained weights=CLIP2025.08 | 72.1 | 65.3 | 88.4 |