ResearchDatasetsAccuracy-based tasksFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsZero-shot downstream reasoning and question answeringAccuracy-based tasks (ARC-E, PIQA, SciQ, HellaSwag, LAMBADA, WinoGrande, BoolQ) zero-shot51.22ARC-E2
Zero-shot downstream reasoning and question answeringAccuracy-based tasks (ARC-E, PIQA, SciQ, HellaSwag, LAMBADA, WinoGrande, BoolQ) zero-shot51.22ARC-E2