Backdoor Detection on Sumo-Humans
88.1AUROCPolicyGuard
Evaluation Results
| Method | Links | |
|---|---|---|
| PolicyGuardStrategy=Original2026.06 | 88.1 | |
| PolicyGuardStrategy=Hard-coded2026.06 | 88.1 | |
| SHINEStrategy=Hard-coded2026.06 | 82.6 | |
| SHINEStrategy=Original2026.06 | 81.6 | |
| PolicyCleanseStrategy=Original2026.06 | 70.8 | |
| PolicyCleanseStrategy=Hard-coded2026.06 | 50.3 |