ResearchDatasetsMMLU & SuperGLUEFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsMultitask Language UnderstandingMMLU & SuperGLUE BoolQ, MultiRC, RTE, WiC (test)51.6MMLU Accuracy6