ResearchTasksConversational AlignmentConversational AlignmentFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedAlpaca-EvalsODRPO Gini-Med (GRPO)58.12Alpaca-Evals Score12May 14, 2026