ResearchDatasetsMNLI, SCIQ, LAMBADA, HellaSwag, ARC, MMLUMNLI, SCIQ, LAMBADA, HellaSwag, ARC, MMLUFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsDownstream Task EvaluationMNLI, SCIQ, LAMBADA, HellaSwag, ARC, MMLU0.385MNLI Acc2