ResearchTasksSubject Knowledge EvaluationFor AI assistantsMarkdownContent indexSubject Knowledge EvaluationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedMMLUPractical (R̄_FPO)48.7MMLU Overall Accuracy3May 7, 2026