ResearchTasksModel Ranking Correlation AnalysisFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedSEED-BenchM2IRT0.94Spearman Correlation (10% subset)7Mar 4, 2026MathVistaM3IRT0.92Spearman Correlation (10% subset)7Mar 4, 2026MMMUM2IRT0.96Spearman Correlation (10% Subset)7Mar 4, 2026