ResearchDatasetsASSE-SafetyFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsTrajectory-level safety evaluationASSE-Safety (test)81.1Accuracy20Agent Safety AuditingASSE-Safety85.4Accuracy13Binary safe/unsafe classificationASSE-Safety (test)67.4Accuracy4