ResearchDatasetsMATH-PerturbMATH-PerturbFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsMathematical Problem SolvingMATH-Perturb MP-simple72.8Exact Match Accuracy12Mathematical ReasoningMath-Perturb79.1Math-P Hard Score5