Image Classification on ImageNet (Clean Accuracy, Robust Accuracy)
71.5Clean AccuracyTTE
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| TTEBackbone=CLIP-L/14, Method category=Test-Time Defense2026.01 | 71.5 | 32.4 | |
| R-TPTBackbone=CLIP-L/14, Method category=Test-Time Defense2026.01 | 71.2 | 60 | |
| CLIPBackbone=CLIP-L/14, Method category=Original2026.01 | 69.9 | 0.6 | |
| TeCoABackbone=CLIP-L/14, Method category=Adversarial Fine-tuning2026.01 | 69.2 | 67.5 | |
| Anti-AdvBackbone=CLIP-L/14, Method category=Test-Time Defense2026.01 | 68 | 37.1 | |
| CSRBackbone=CLIP-L/14, Method category=Test-Time Defense2026.01 | 68 | 64.1 | |
| HDBackbone=CLIP-L/14, Method category=Test-Time Defense2026.01 | 66.4 | 10.1 | |
| LPFBackbone=CLIP-L/14, Method category=Test-Time Defense2026.01 | 64.2 | 43.2 | |
| FAREBackbone=CLIP-L/14, Method category=Adversarial Fine-tuning2026.01 | 63.9 | 61.5 | |
| TTCBackbone=CLIP-L/14, Method category=Test-Time Defense2026.01 | 54.7 | 35.8 |