ResearchDatasetsMMLU, ARC-c, ARC-e, CSQA, OBQA, PIQA, SIQA, WinograndeFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsMultiple-choice Question AnsweringMMLU, ARC-c, ARC-e, CSQA, OBQA, PIQA, SIQA, Winogrande zero-shot30.81MMLU Accuracy (zero-shot)9
Multiple-choice Question AnsweringMMLU, ARC-c, ARC-e, CSQA, OBQA, PIQA, SIQA, Winogrande zero-shot30.81MMLU Accuracy (zero-shot)9