ResearchTasksPairwise PreferencePairwise PreferenceFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedGenAI Bench (test)HPSv3++76.3Accuracy18Jun 15, 2026HPD v3 (test)HPSv3++86.7Accuracy18Jun 15, 2026