ResearchDatasetsAverage (GSM8K, MATH, MMLU-MATH, MATHQA)Average (GSM8K, MATH, MMLU-MATH, MATHQA)FollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsMathematical ReasoningAverage (GSM8K, MATH, MMLU-MATH, MATHQA) (test)51.4Accuracy6