ResearchDatasetsMT_BenchMT_BenchFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsGeneral Utility EvaluationMT_Bench82.7Agreement Rate33Multi-turn Dialogue EvaluationMT_Bench CodaSet OOD (test)98.16Performance (%)16