ResearchTasksPreference LabelingFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedSummarizationCurriculum-RLAIF89Preference Labeling Accuracy8Apr 21, 2026Anthropic HelpfulnessCurriculum-RLAIF81Preference Labeling Accuracy8Apr 21, 2026Anthropic HarmlessnessCurriculum-RLAIF77Preference Labeling Accuracy8Apr 21, 2026