ResearchTasksResponse Repetition Rate EvaluationResponse Repetition Rate EvaluationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedJailbreakBench (JBB) (test)Qwen3-Max7MS_R (Repetition Rate)5Mar 12, 2026HarmBench (test)Qwen3 1.7B26.5MS_R3Mar 12, 2026AdvBench Adv (test)Qwen3 4B43.5MS_R3Mar 12, 2026