ResearchDatasetsHarmBench, TruthfulQA, and BBQHarmBench, TruthfulQA, and BBQFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsSafety EvaluationHarmBench, TruthfulQA, and BBQ91.4Safety Score16