ResearchDatasetsArena-Expert, HelpSteer, HH-RLHF, and UltraFeedbackFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsPreference PredictionArena-Expert-5K, HelpSteer3, HH-RLHF, and UltraFeedback (held-out)70.5Accuracy42
Preference PredictionArena-Expert-5K, HelpSteer3, HH-RLHF, and UltraFeedback (held-out)70.5Accuracy42