Medical Image Segmentation on Site E
80.5Dice ScoreMCDRL
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| MCDRLBackbone=ViT-L/14, Pre-trained weights=CLIP2025.08 | 80.5 | 73.3 | 93.7 | |
| MCDRLBackbone=ViT-B/16, Pre-trained weights=CLIP2025.08 | 78.8 | 71.8 | 92.9 | |
| BiomedCoOpBackbone=ViT-L/14, Pre-trained weights=CLIP2025.08 | 78.3 | 71.4 | 92.7 | |
| MCDRLBackbone=ResNet-50, Pre-trained weights=CLIP2025.08 | 77.3 | 70.4 | 92 | |
| StyLIPBackbone=ViT-L/14, Pre-trained weights=CLIP2025.08 | 77.2 | 70.1 | 92 | |
| BiomedCoOpBackbone=ViT-B/16, Pre-trained weights=CLIP2025.08 | 76.8 | 70.1 | 91.8 | |
| BiomedCoOpBackbone=ResNet-50, Pre-trained weights=CLIP2025.08 | 75.6 | 69.1 | 90.9 | |
| StyLIPBackbone=ViT-B/16, Pre-trained weights=CLIP2025.08 | 75.6 | 68.8 | 91.1 | |
| StyLIPBackbone=ResNet-50, Pre-trained weights=CLIP2025.08 | 74.1 | 67.5 | 90.1 | |
| BaselineBackbone=ViT-L/14, Pre-trained weights=CLIP2025.08 | 73.5 | 67.1 | 90.2 | |
| BaselineBackbone=ViT-B/16, Pre-trained weights=CLIP2025.08 | 72.1 | 65.6 | 89.3 | |
| BaselineBackbone=ResNet-50, Pre-trained weights=CLIP2025.08 | 70.4 | 64.2 | 88.1 |