ResearchTasksContinuous-risk assessmentFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedSIA-Bench SkillJect-derived threat slice (ood)Contextual Invocation Audit40.5HR-AUPRC2Apr 14, 2026SIA-Bench SkillJect-derived threat slice (test)Static Denylist32.7HR-AUPRC2Apr 14, 2026SIA-Bench IPIGuard replay (ood)Contextual Invocation Audit40.5HR-AUPRC2Apr 14, 2026SIA-Bench IPIGuard replay (test)Static Denylist32.7HR-AUPRC2Apr 14, 2026