ResearchTasksOpen-ended arithmetic reasoningOpen-ended arithmetic reasoningFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedMATH (test)Critical-CoT96.9TDR6Apr 14, 2026GSM8K (test)Critical-CoT99.3TDR6Apr 14, 2026