List your company
ResearchFormerly WizWand

Mathematical Reasoning on AMC 2023 (average@16 accuracy)

70.9Average@16 AccuracyPROF-BOTH
8.18824.46940.7557.031Sep 3, 2025
Updated 5mo ago

Evaluation results

MethodLinks
PROF-BOTHModel=Qwen2.5-Math-7B-base, Temperature=1.0, Evaluation Protocol=average@162025.0970.9
GRPOModel=Qwen2.5-Math-7B-base, Temperature=1.0, Evaluation Protocol=average@162025.0964.4
PROF-POSModel=Qwen2.5-Math-7B-base, Temperature=1.0, Evaluation Protocol=average@162025.0964.2
BlendModel=Qwen2.5-Math-7B-base, Temperature=1.0, Evaluation Protocol=average@162025.0958
PROF-POSModel=Qwen2.5-Math-1.5B-base, Temperature=1.0, Evaluation Protocol=average@162025.0950.3
PROF-BOTHModel=Qwen2.5-Math-1.5B-base, Temperature=1.0, Evaluation Protocol=average@162025.0949.1
GRPOModel=Qwen2.5-Math-1.5B-base, Temperature=1.0, Evaluation Protocol=average@162025.0944.5
BlendModel=Qwen2.5-Math-1.5B-base, Temperature=1.0, Evaluation Protocol=average@162025.0942.5
BaseModel=Qwen2.5-Math-7B-base, Temperature=1.0, Evaluation Protocol=average@162025.0930
PROF-POSModel=LLaMA-3.2-3B-instruct, Temperature=1.0, Evaluation Protocol=average@162025.0928.6
GRPOModel=LLaMA-3.2-3B-instruct, Temperature=1.0, Evaluation Protocol=average@162025.0925.6
PROF-BOTHModel=LLaMA-3.2-3B-instruct, Temperature=1.0, Evaluation Protocol=average@162025.0923.9
BaseModel=Qwen2.5-Math-1.5B-base, Temperature=1.0, Evaluation Protocol=average@162025.0923.6
BlendModel=LLaMA-3.2-3B-instruct, Temperature=1.0, Evaluation Protocol=average@162025.0917.2
BaseModel=LLaMA-3.2-3B-instruct, Temperature=1.0, Evaluation Protocol=average@162025.0910.6