Adversarial and Jailbreaking Attack Detection on XSTest
0.984AUROC
GradSafe
Updated 28d ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.05 | 0.984 | - | |
2026.05 | 0.97 | - | |
2026.05 | 0.95 | - | |
2026.05 | 0.949 | - | |
2026.05 | 0.948 | - | |
2026.05 | 0.946 | - | |
2026.05 | 0.942 | - | |
2026.05 | 0.932 | - | |
2026.05 | 0.93 | - | |
2026.05 | 0.893 | - | |
2026.02 | 0.8418 | 0.781 | |
2026.02 | 0.8257 | 0.9667 | |
2026.02 | 0.812 | 1 | |
2026.02 | 0.8021 | 0.781 | |
2026.02 | 0.8007 | 0.8714 | |
2026.02 | 0.7932 | 0.7381 | |
2026.02 | 0.7929 | 1 | |
2026.02 | 0.7906 | 0.9238 | |
2026.05 | 0.786 | - | |
2026.02 | 0.7843 | 1 | |
2026.05 | 0.716 | - | |
2026.02 | 0.6794 | 0.8143 | |
2026.02 | 0.6542 | 0.9095 | |
2026.02 | 0.6308 | 0.9952 | |
2026.05 | 0.595 | - | |
2026.02 | 0.58 | 0.8762 | |
2026.05 | 0.57 | - | |
2026.05 | 0.481 | - | |
2026.02 | 0.373 | 1 | |
2026.02 | 0.3345 | 1 | |
2026.02 | 0.2803 | 0.9952 | |
2026.02 | 0.2373 | 1 | |
2026.02 | 0.2298 | 1 | |
2026.02 | 0.2144 | 1 | |
2026.02 | 0.1565 | 1 |