| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| CIFAR-10 (test) | FGSM | Attack Success Rate (ASR)47.7 | 76 | 1mo ago | |
| CIFAR-10 | Muon | PGD Accuracy (L-inf)49.71 | 48 | 1mo ago | |
| CIFAR-100 (test) | FAT | Natural Acc66.74 | 46 | 1mo ago | |
| CIFAR-10 | HICAT | Clean Accuracy89.85 | 40 | 1mo ago | |
| ImageNet 1k (test) | RVT-B* | FGSM Robustness53 | 34 | 4mo ago | |
| CIFAR-10 | Proposed method (FGSM) | FGSM Robust Accuracy99.2 | 30 | 2mo ago | |
| CIFAR100 1000 images (val) | SA | Clean Acc70.25 | 24 | 4mo ago | |
| CIFAR10 1000 images (test) | Square Attack (SA) | Robust Accuracy72.3 | 24 | 4mo ago | |
| GCG | GCG Rate0.13 | 21 | 4mo ago | ||
| CIFAR-100 LT (test) | AT-BSL-AuA | Clean Accuracy50.66 | 20 | 4mo ago | |
| CIFAR-100 | HICAT | Clean Accuracy66.02 | 18 | 1mo ago | |
| GTSRB (test) | FGSM-Neuro-Symbolic | FGSM Accuracy74 | 18 | 4mo ago | |
| AutoDAN | ASR0 | 18 | 4mo ago | ||
| PAIR | ASR26 | 18 | 4mo ago | ||
| ImageNet 1k (val) | BG_Random | Accuracy13.4 | 18 | 4mo ago | |
| CIFAR-100 | Cutout | Final Auto-Attack Accuracy98.5 | 16 | 4mo ago | |
| Tiny ImageNet AA | HEM- | DAR31.42 | 15 | 1mo ago | |
| CIFAR100 AA | DICE | DAR61.32 | 15 | 1mo ago | |
| CIFAR10 AA | DICE | DAR12.34 | 15 | 1mo ago | |
| MNIST AA | HEM- | DAR85.04 | 15 | 1mo ago | |
| ImageNet-1K | DICE | DAR84.76 | 15 | 1mo ago | |
| ImageNet sr=90% (val) | AT | Clean Accuracy83.29 | 14 | 4mo ago | |
| CIFAR-10 | PARDEF | Attack Success Rate9.76 | 14 | 1mo ago | |
| Skill-Composed 2k queries (test) | Explicit Refusals Count169 | 12 | 3mo ago | ||
| WildJailbreak 2k queries (test) | AUTOSKILL | Number of Explicit Refusals498 | 12 | 3mo ago |
