ResearchTasksMulti-subject KnowledgeMulti-subject KnowledgeFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedMMLUBaseline79.7Accuracy16May 13, 2026MMLU-ReduxReLIFT67Accuracy9Mar 12, 2026