ResearchDatasetsMMLU-ReduxFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsLanguage UnderstandingMMLU-Redux Generative87.8Humanities Accuracy16Knowledge EvaluationMMLU-Redux 2.0 (Continual)33.49Accuracy6Knowledge EvaluationMMLU-Redux 2.0 (Original)42.03Accuracy6Query RoutingMMLU-Redux OOD52.5CPT (80%)4