ResearchDatasetsPIQA, WinoGrande, HellaSwag, BoolQ, SocialIQA, OpenBookQAPIQA, WinoGrande, HellaSwag, BoolQ, SocialIQA, OpenBookQAFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsCommonsense ReasoningPIQA, WinoGrande, HellaSwag, BoolQ, SocialIQA, OpenBookQA90.2PIQA Accuracy7