ResearchTasksCommonsense and Language ReasoningCommonsense and Language ReasoningFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedCommonsense and Language Aggregate: HellaSwag, Winogrande, BoolQ (test)MSFT70.6Accuracy6Mar 24, 2026