Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Policy Violation Enforcement on MedAgentBench (Original)

0Unsafe Rate

Hong et al. (Manual Guardrails)

-1.568.9719.530.03Jun 25, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
0-
2026.06
23-
39-
2026.06
-30.3
2026.06
-57.7
2026.06
-55.7