ResearchDatasetsHH and UFFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsPreference AlignmentHH and UF Out-of-Domain (test)58.7OOD Win Rate10Preference AlignmentHH and UF In-Domain (test)68.4Win Rate10