ResearchDatasetsWinogrande, OBQA, Hellaswag, Boolq, ARC, RTEWinogrande, OBQA, Hellaswag, Boolq, ARC, RTEFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsZero-shot EvaluationWinogrande, OBQA, Hellaswag, Boolq, ARC, RTE (test)76.07Winogrande Accuracy9