ResearchDatasetsOmniMathOmniMathFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsMathematical ReasoningOmniMath (test)55.1Accuracy20Mathematical ReasoningOMNIMATH random subset of 128 samples10.93Top-1 Accuracy12Mathematical ReasoningOmniMath30Accuracy6Mathematical ReasoningOmniMath (train)50.1Training Dataset (%)3Math ReasoningOmniMath56.5BA @20 Accuracy2