ResearchDatasetsSafeDialBenchSafeDialBenchFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsSafety Dialogue EvaluationSafeDialBench8.42Score33Safety DetectionSafeDialBench (full)99Recall12Jailbreak DetectionSafeDialBench0.912AUC9Safety dialogue evaluationSafeDialBench61.33Normalized Score5Unsafe-input detectionSafeDialBench EN99.07Recall2