List your company

Mathematical Reasoning on AIME 2025 (Acc, ECE, Brier, AUROC)

0.211AccuracyCAR (MSCR)
0.197480.200990.20450.20801Jan 12, 2026

Evaluation results

MethodLinks
CAR (MSCR)Model=Qwen2.5-3B2026.010.2110.5190.410.695
Vanilla SimpleTIRModel=Qwen2.5-3B2026.010.1980.6870.6320.498