ResearchDatasetsPKU-HH-SafeRLHFPKU-HH-SafeRLHFFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsReward AccuracyCleaned-PKU-HH-SafeRLHF (test)91.8Reward Accuracy15