ResearchDatasetsWe-MathWe-MathFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsMathematical ReasoningWe-Math mini (test)86.4Accuracy31Math ReasoningWe-Math85.6Avg Pass@826Multi-step mathematical reasoningWe-Math (test)72.8S1 Score20Mathematical & Geometric ReasoningWe-Math77.7Accuracy@816