ResearchTasksHuman feedback evaluation consistencyFor AI assistantsMarkdownContent indexHuman feedback evaluation consistencyFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedRichHF-18KUnifiedReward_Q(8B)40.4Pearson-r11Apr 30, 2026