Image Classification on ImageNet OOD Suite (test)
72.04Accuracy (ImageNet-V2)CLIP ViT-B/16 + ED
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| CLIP ViT-B/16 + EDBackbone=ViT-B/16, Regularization=Effective Degree (ED)2026.05 | 72.04 | 66.3 | 39.81 | 47.53 | 50.76 | 55.29 | |
| CLIP ViT-B/16Backbone=ViT-B/162026.05 | 70.89 | 65.32 | 36.63 | 45.45 | 50.14 | 53.69 | |
| CLIP ViT-B/32 + EDBackbone=ViT-B/32, Regularization=Effective Degree (ED)2026.05 | 65.37 | 58.28 | 22.03 | 40.46 | 40.41 | 45.31 | |
| CLIP ViT-B/32Backbone=ViT-B/322026.05 | 64.21 | 56.82 | 20.48 | 39.08 | 39.62 | 44.04 |