ResearchDatasetsMMLU, NaturalQuestions, TriviaQAMMLU, NaturalQuestions, TriviaQAFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsKnowledgeMMLU, NaturalQuestions, TriviaQA60.1MMLU (5-shot)13