ResearchTasksPolicy violation enforcementFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedMedAgentBench (Adversarial)Hong et al. (Manual Guardrails)0Unsafe Rate (Hong et al.)3Jun 26, 2026MedAgentBench (Original)Hong et al. (Manual Guardrails)0Unsafe Rate3Jun 26, 2026