ResearchDatasetsPIQA, HellaSwag, WinoGrande, ARCPIQA, HellaSwag, WinoGrande, ARCFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsCommonsense reasoningPIQA, HellaSwag, WinoGrande, ARC79.4PIQA Accuracy4