ResearchTasksHarmful input detectionFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedSafety Evaluation Suite (AEG, AEG2, HB-A, HB-V, OAI, SAL, SST, Toxic, WGT, WJB)WildGuard89.8AEG Score18Jun 29, 2026JailbreakBench ConvReGA100MCA Accuracy7Apr 21, 2026RepE-HarmfulReGA100MCA Accuracy7Apr 21, 2026JailbreakBenchReGA100Accuracy (MCA)7Apr 21, 2026HarmBenchReGA100MCA Accuracy7Apr 21, 2026
Safety Evaluation Suite (AEG, AEG2, HB-A, HB-V, OAI, SAL, SST, Toxic, WGT, WJB)WildGuard89.8AEG Score18Jun 29, 2026