ResearchDatasetsGSM8K, CommonSense, BoolQ, ARC, and HellaSwagGSM8K, CommonSense, BoolQ, ARC, and HellaSwagFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsReasoning and MathGSM8K, CommonSense, BoolQ, ARC Challenge, and HellaSwag87.8Average Accuracy9