| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| XSTest | DUOGUARD | AUROC0.8418 | 20 | 4d ago | |
| MaliciousInstruct | AUROC0.8825 | 20 | 4d ago | ||
| JailbreakBench | T3-OCSVM | AUROC0.8622 | 20 | 4d ago | |
| HarmBench | LLAMAGUARD3-1B-LOGITS | AUROC0.8887 | 20 | 4d ago | |
| BeaverTails | DUOGUARD | AUROC0.8525 | 20 | 4d ago | |
| AdvBench | T3-GMM | AUROC0.9675 | 20 | 4d ago |