ResearchTasksCode Reasoning (Input Prediction)FollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedCruxEval-I50% CodeEnhance + 50% Nemotron-RW35.7CX-I10Jun 11, 2026CRUXEval-I 1-shotERNIE 5.0-Base79.75Accuracy3Feb 26, 2026