List your company
ResearchFormerly WizWand

Mathematical Reasoning on OMNIMATH (128-sample random subset)

10.93Top-1 AccuracyBeam-Search
5.24126.71818.1959.6719Oct 7, 2025
Updated 7mo ago

Evaluation results

MethodLinks
Beam-SearchSelection=Argmax, Scoring=PRM, Base Model=Qwen2.5-7B-Instruct2025.1010.93
ePFSelection=Argmax, Scoring=PRM, Base Model=Qwen2.5-7B-Instruct2025.1010.93
Best-of-NSelection=Argmax, Scoring=ORM, Base Model=Qwen2.5-1.5B-Instruct2025.1010.15
ePFSelection=Argmax, Scoring=PRM, Base Model=Qwen2.5-1.5B-Instruct2025.1010.15
PFSelection=Argmax, Scoring=PRM, Base Model=Qwen2.5-7B-Instruct2025.1010.15
Beam-SearchSelection=Argmax, Scoring=PRM, Base Model=Qwen2.5-1.5B-Instruct2025.109.37
Self-ConsistencySelection=MV, Base Model=Qwen2.5-7B-Instruct2025.109.37
Best-of-NSelection=Argmax, Scoring=ORM, Base Model=Qwen2.5-7B-Instruct2025.109.37
PFSelection=Argmax, Scoring=PRM, Base Model=Qwen2.5-1.5B-Instruct2025.108.59
Base SamplingBase Model=Qwen2.5-7B-Instruct2025.108.59
Self-ConsistencySelection=MV, Base Model=Qwen2.5-1.5B-Instruct2025.107.03
Base SamplingBase Model=Qwen2.5-1.5B-Instruct2025.105.46