Image Classification on FGVC zero-shot
26.8Clean AccuracyVanilla FT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Vanilla FTBackbone=ViT-B/16, Training=ImageNet, Evaluation=Zero-shot2026.03 | 26.8 | 0 | |
| GRACEBackbone=ViT-B/16, Training=ImageNet, Evaluation=Zero-shot2026.03 | 26.4 | 6.2 | |
| TPGMBackbone=ViT-B/16, Training=ImageNet, Evaluation=Zero-shot2026.03 | 25.1 | 0 | |
| SPDBackbone=ViT-B/16, Training=ImageNet, Evaluation=Zero-shot2026.03 | 24.7 | 0 | |
| FLYPBackbone=ViT-B/16, Training=ImageNet, Evaluation=Zero-shot2026.03 | 24.3 | 0 | |
| WISE-FTBackbone=ViT-B/16, Training=ImageNet, Evaluation=Zero-shot2026.03 | 22.5 | 0 | |
| CLIPBackbone=ViT-B/16, Training=ImageNet, Evaluation=Zero-shot2026.03 | 18.96 | 0 | |
| FAREBackbone=ViT-B/16, Training=ImageNet, Evaluation=Zero-shot2026.03 | 15.3 | 2.1 | |
| PMG-AFTBackbone=ViT-B/16, Training=ImageNet, Evaluation=Zero-shot2026.03 | 10.25 | 4.85 | |
| LAATBackbone=ViT-B/16, Training=ImageNet, Evaluation=Zero-shot2026.03 | 9.2 | 4.1 | |
| TeCoABackbone=ViT-B/16, Training=ImageNet, Evaluation=Zero-shot2026.03 | 7.81 | 3.13 |