Multiclass image-text classification on KID2
48.7Macro F1CapCLIP-M (ViT-L/14)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CapCLIP-M (ViT-L/14)Zero-shot=true, Backbone=ViT-L/142026.05 | 48.7 | 60 | |
| CapCLIP-M (ViT-B/16)Zero-shot=true, Backbone=ViT-B/162026.05 | 44.2 | 55.1 | |
| CapCLIP-MIX (ViT-L/14)Zero-shot=true, Backbone=ViT-L/142026.05 | 38.9 | 44.8 | |
| CapCLIP-MIX (ViT-B/16)Zero-shot=true, Backbone=ViT-B/162026.05 | 35.9 | 47.9 | |
| CapCLIP-B (ViT-L/14)Zero-shot=true, Backbone=ViT-L/142026.05 | 33.1 | 48.8 | |
| CapCLIP-B (ViT-B/16)Zero-shot=true, Backbone=ViT-B/162026.05 | 32.8 | 46 | |
| BioMedCLIPZero-shot=true2026.05 | 31.6 | 43.7 | |
| MedImageInsightZero-shot=true2026.05 | 25.4 | 32.5 | |
| CLIP-OpenAIZero-shot=true2026.05 | 23.5 | 32.9 | |
| MedCLIPZero-shot=true2026.05 | 20.4 | 28.7 |