ResearchDatasetsHEx-PHI and AdvBenchHEx-PHI and AdvBenchFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsHarmful Response GenerationHEx-PHI and AdvBench (test)96.86ASR7Harmfulness EvaluationHEx-PHI & AdvBench Primary1.15Primary Average6