Label Prediction on WASP
90.6AccuracySAFEPRED
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| SAFEPREDType=Predictive2026.02 | 90.6 | 4.5 | |
| HarmonyGuardType=Reactive2026.02 | 89.5 | 5.2 | |
| gpt-oss-safeguard-20bType=Reactive2026.02 | 89.4 | 2.5 | |
| GuardAgentType=Reactive2026.02 | 78.9 | 6.2 |