Image Classification on Oxford-IIIT Flowers-102 (Adversarial Evaluation)
60.66Top-1 Accuracy (Clean)WISE-FT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| WISE-FTBackbone=ViT-B/32, Pre-training Dataset=ImageNet, Evaluation Protocol=Zero-shot2026.03 | 60.66 | 0 | |
| CLIPBackbone=ViT-B/32, Evaluation Protocol=Zero-shot2026.03 | 60.18 | 0 | |
| GRACEBackbone=ViT-B/32, Pre-training Dataset=ImageNet, Evaluation Protocol=Zero-shot2026.03 | 60.05 | 7 | |
| SPDBackbone=ViT-B/32, Pre-training Dataset=ImageNet, Evaluation Protocol=Zero-shot2026.03 | 57.7 | 0 | |
| TPGMBackbone=ViT-B/32, Pre-training Dataset=ImageNet, Evaluation Protocol=Zero-shot2026.03 | 57.18 | 0 | |
| Vanilla FTBackbone=ViT-B/32, Pre-training Dataset=ImageNet, Evaluation Protocol=Zero-shot2026.03 | 55.42 | 0 | |
| FLYPBackbone=ViT-B/32, Pre-training Dataset=ImageNet, Evaluation Protocol=Zero-shot2026.03 | 50.18 | 0 | |
| PMG-AFTBackbone=ViT-B/32, Pre-training Dataset=ImageNet, Evaluation Protocol=Zero-shot2026.03 | 33.4 | 7.3 | |
| FAREBackbone=ViT-B/32, Pre-training Dataset=ImageNet, Evaluation Protocol=Zero-shot2026.03 | 31.72 | 5.3 | |
| LAATBackbone=ViT-B/32, Pre-training Dataset=ImageNet, Evaluation Protocol=Zero-shot2026.03 | 28.9 | 6.1 | |
| TeCoABackbone=ViT-B/32, Pre-training Dataset=ImageNet, Evaluation Protocol=Zero-shot2026.03 | 25.59 | 6.6 |