ResearchDatasetsAlpaca and AdvBenchAlpaca and AdvBenchFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsSafety DetectionAlpaca and AdvBench Conversation Level (test)100MCA Accuracy7Safety DetectionAlpaca and AdvBench Prompt Level (test)100Accuracy (MCA)7