ResearchDatasetsGSM8k, SAT-Math, & MATHGSM8k, SAT-Math, & MATHFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsMath problem solvingGSM8k, SAT-Math, & MATH OpenCompass AGIEval sampled (test)32.22GSM8k Accuracy4