ResearchDatasetsAverage (2WikiMultiHop, MMLU, GSM8k)FollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsMulti-task ReasoningAverage (2WikiMultiHop, MMLU, GSM8k) (in-distribution)75.2Accuracy29