Adversarial and Jailbreaking Attack Detection on AdvBench
0.9675AUROC
T3-GMM
Updated 4d ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.02 | 0.9675 | 0.1577 | |
2026.02 | 0.9578 | 0.1731 | |
2026.02 | 0.8908 | 0.8269 | |
2026.02 | 0.8894 | 1 | |
2026.02 | 0.8822 | 1 | |
2026.02 | 0.867 | 0.6654 | |
2026.02 | 0.8658 | 1 | |
2026.02 | 0.8241 | 0.9327 | |
2026.02 | 0.811 | 0.75 | |
2026.02 | 0.7895 | 0.9558 | |
2026.02 | 0.7814 | 0.9942 | |
2026.02 | 0.5894 | 0.9827 | |
2026.02 | 0.5689 | 0.975 | |
2026.02 | 0.4461 | 0.9962 | |
2026.02 | 0.3989 | 0.9981 | |
2026.02 | 0.3681 | 0.9942 | |
2026.02 | 0.3575 | 0.9981 | |
2026.02 | 0.334 | 0.9981 | |
2026.02 | 0.2963 | 1 | |
2026.02 | 0.2737 | 0.9981 |