ResearchDatasetsCoTFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsInstruction TuningCoT70.55Reasoning Score20CS research tasksCoT 200-prompt (held-out test)77.4Mean@33Prompting algorithm implementationCoT-SC35AST Size3