ResearchBenchmarksModel Ranking Correlation Analysis on MMMU (10%-50% subsets)Follow0.96Spearman Correlation (10% Subset)M2IRT0.1280.3440.560.776Mar 3, 2026Evaluation ResultsMethodMethodLinksSpearman Correlation (10% Subset)Spearman Correlation (20% Subset)Spearman Correlation (30% Subset)Spearman Correlation (40% Subset)Spearman Correlation (50% Subset)M2IRT2026.030.960.960.960.970.96M3IRT2026.030.890.920.920.930.93FlashEval2026.030.790.790.790.80.8Random2026.030.770.80.820.820.81IRT2026.030.50.360.380.350.12TinyBenchmarks2026.030.430.50.560.610.67MIRT2026.030.160.330.470.610.69