ResearchTasksRule-based detection performanceRule-based detection performanceFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedBenign Web baselinedeterministic synthesis function0Union Match Rate1Jun 5, 2026Benign LLM baselinedeterministic synthesis function2Union Match Rate1Jun 5, 2026HarmBench (held-out)deterministic synthesis function7Union Match1Jun 5, 2026AdvBench (held-out)deterministic synthesis function15Union Match Rate1Jun 5, 2026