List your company

Mathematical Reasoning on Average (AIME23, AMC23, MATH500)

75.2Majority Score (Avg)PAEC
0.73620.06839.458.732Jun 7, 2026

Evaluation results

MethodLinks
PAECModel=Qwen3-4B-Base2026.0675.262
GRPOModel=Qwen3-4B-Base2026.0671.960.4
Global Entropy RegModel=Qwen3-4B-Base2026.0669.759.5
BaseModel=Qwen3-4B-Base2026.0666.757.6
PAECModel=Llama3.1-8B2026.0610.96.7
GRPOModel=Llama3.1-8B2026.067.95.8
Global Entropy RegModel=Llama3.1-8B2026.067.45.5
BaseModel=Llama3.1-8B2026.063.61.6