List your company

Safety Monitoring on BeaverTails

19.6ASR @ 1% FPRQwenGuard
17.81629.85841.953.942Mar 24, 2026

Evaluation results

MethodLinks
QwenGuard2026.0319.684.89
ActProbe2026.0327.2985.94
LlamaGuard2026.0332.674.54
ActWM2026.0333.287.79
Base2026.0364.2—