ResearchDatasetsSkill-Unseen MethodFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsInput prompt classificationSkill-Unseen Method (70B)100Accuracy5Input prompt classificationSkill-Unseen Method 8B99.93Accuracy5Input prompt classificationSkill-Unseen Method 4B100Accuracy5