List your company

Mathematical Reasoning on MATH (Accuracy (%))

90.8Accuracy (%)Pass@k
69.16874.78480.486.016Jun 1, 2026

Evaluation results

MethodLinks
Pass@kk=16, Model=Qwen2.5-7B2026.0690.8
PRM-72B guided searchk=16, L=20, Base Model=Qwen2.5-7B, Guidance Model=Qwen2.5-72B2026.0684
CGSk=16, L=20, Base Model=Qwen2.5-7B, Guidance Model=Qwen2.5-72B2026.0681.8
Large greedyModel=Qwen2.5-72B2026.0680.6
LGSk=16, L=20, Base Model=Qwen2.5-7B, Guidance Model=Qwen2.5-72B2026.0680.2
Borda count (small)k=16, Model=Qwen2.5-7B2026.0678
Borda count (large)k=16, Model=Qwen2.5-72B2026.0677.4
Majority@kk=16, Model=Qwen2.5-7B2026.0677.2
Self-Certainty (large)k=16, Model=Qwen2.5-72B2026.0675.8
Best-of-Nk=16, Model=Qwen2.5-7B2026.0675.8
Self-Certainty (small)k=16, Model=Qwen2.5-7B2026.0672.8
Small greedyModel=Qwen2.5-7B2026.0670