ResearchDatasetsAI2 OLMESFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsZero-shot EvaluationAI2 OLMES zero-shot35.96ScienceQA15General Language UnderstandingAI2 OLMES (test)40.4ARC-E Accuracy4