ResearchDatasetsMMMU, MathVision, MathVerse, DynaMath, WeMathFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsFirst Incorrect Step IdentificationMMMU, MathVision, MathVerse-VO, DynaMath, WeMath Overall26.4FISI F16
First Incorrect Step IdentificationMMMU, MathVision, MathVerse-VO, DynaMath, WeMath Overall26.4FISI F16