List your company

Mathematical Reasoning on MATH first 1,000 samples (test)

86.8Exact Match AccuracyMGRS
77.9680.25582.5584.845Nov 28, 2025

Evaluation results

MethodLinks
MGRSBackbone=GPT-4o-mini2025.1186.8
AoTBackbone=GPT-4o-mini2025.1183.6
AFlowBackbone=GPT-4o-mini2025.1183
FoTBackbone=GPT-4o-mini, n=82025.1182.5
CoT-SCBackbone=GPT-4o-mini, n=52025.1181.8
Self-refineBackbone=GPT-4o-mini2025.1178.7
CoTBackbone=GPT-4o-mini2025.1178.3