ResearchDatasetsWJBWJBFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsHarmful RefusalWJB0.2ASR26Harmful prompt detectionWJB97.55F1 Score17Jailbreak DetectionWJB93.26ACC13Safety EvaluationWJB7.5ASR (%)5