List your company

Model Ranking Correlation Analysis on MathVista

0.92Spearman Correlation (10% subset)M3IRT
0.56640.65820.750.8418Mar 3, 2026

Evaluation results

MethodLinks
M3IRT2026.030.920.940.930.930.93
FlashEval2026.030.890.910.910.910.91
Random2026.030.890.910.920.930.93
M2IRT2026.030.810.840.880.910.93
IRT2026.030.810.880.910.930.94
TinyBenchmarks2026.030.790.840.860.880.88
MIRT2026.030.580.670.720.760.81