List your company
ResearchFormerly WizWand

Mathematical Reasoning on Olympiad Bench (average@16 accuracy)

47.8Average@16 AccuracyPROF-BOTH
4.43215.69126.9538.209Sep 3, 2025
Updated 5mo ago

Evaluation results

MethodLinks
PROF-BOTHModel=Qwen2.5-Math-7B-base, Temperature=1.0, Evaluation Protocol=average@162025.0947.8
GRPOModel=Qwen2.5-Math-7B-base, Temperature=1.0, Evaluation Protocol=average@162025.0945.5
BlendModel=Qwen2.5-Math-7B-base, Temperature=1.0, Evaluation Protocol=average@162025.0945
PROF-POSModel=Qwen2.5-Math-7B-base, Temperature=1.0, Evaluation Protocol=average@162025.0945
PROF-POSModel=Qwen2.5-Math-1.5B-base, Temperature=1.0, Evaluation Protocol=average@162025.0936.1
PROF-BOTHModel=Qwen2.5-Math-1.5B-base, Temperature=1.0, Evaluation Protocol=average@162025.0936.1
GRPOModel=Qwen2.5-Math-1.5B-base, Temperature=1.0, Evaluation Protocol=average@162025.0933
BlendModel=Qwen2.5-Math-1.5B-base, Temperature=1.0, Evaluation Protocol=average@162025.0931.1
PROF-POSModel=LLaMA-3.2-3B-instruct, Temperature=1.0, Evaluation Protocol=average@162025.0919.8
BaseModel=Qwen2.5-Math-7B-base, Temperature=1.0, Evaluation Protocol=average@162025.0919.2
BaseModel=Qwen2.5-Math-1.5B-base, Temperature=1.0, Evaluation Protocol=average@162025.0919.1
GRPOModel=LLaMA-3.2-3B-instruct, Temperature=1.0, Evaluation Protocol=average@162025.0917.9
PROF-BOTHModel=LLaMA-3.2-3B-instruct, Temperature=1.0, Evaluation Protocol=average@162025.0917.3
BlendModel=LLaMA-3.2-3B-instruct, Temperature=1.0, Evaluation Protocol=average@162025.099.9
BaseModel=LLaMA-3.2-3B-instruct, Temperature=1.0, Evaluation Protocol=average@162025.096.1