ResearchDatasetsHellaSwag, PIQA, SIQA, WinoGrandeFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsCommonsense ReasoningHellaSwag (HS), PIQA, SIQA, WinoGrande (Wino)78.8HellaSwag Accuracy13