ResearchDatasetsCombined benchmarkCombined benchmarkFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsGold-standard ranking agreementcombined benchmark0.865Mean Kendall's Tau-b36Method ranking self-consistencyCombined benchmark M=120 questions0.893Mean Kendall's Tau-b30