ResearchDatasetsArena-Expert, HelpSteer, HH-RLHF, UltraFeedbackArena-Expert, HelpSteer, HH-RLHF, UltraFeedbackFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsRubric DiscoveryArena-Expert-5K, HelpSteer3, HH-RLHF, UltraFeedback cross-source mean4.45St. Score9