ResearchDatasetsMMLU, GSM, HellaSwag, TruthfulQA, ARC, CodeXFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsAggregate Performance EvaluationMMLU, GSM, HellaSwag, TruthfulQA, ARC-C, CodeX5.32Improvement18