ResearchTasksAgent Safety AuditingAgent Safety AuditingFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedASSE-SecurityTRACES97.6Accuracy13May 28, 2026ASSE StrictShieldAgent83.5Accuracy13May 28, 2026ASSE-SafetyTRACES85.4Accuracy13May 28, 2026ATBenchTRACES85.5Accuracy13May 28, 2026