ResearchDatasetsARC, BoolQ, HSwag, LAMBADA, OBQA, PIQA, SocIQA, WinoGrFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsZero-shot evaluationARC-E, BoolQ, HSwag, LAMBADA, OBQA, PIQA, SocIQA, WinoGr (test)60.65ARC-E Accuracy6
Zero-shot evaluationARC-E, BoolQ, HSwag, LAMBADA, OBQA, PIQA, SocIQA, WinoGr (test)60.65ARC-E Accuracy6