List your company

Mathematical Reasoning on AIME26 (Maj@32, Avg@32)

66.7Maj@32PAEC
-2.66815.34133.3551.359Jun 7, 2026

Evaluation results

MethodLinks
PAECModel=Qwen3-4B-Base2026.0666.743.2
GRPOModel=Qwen3-4B-Base2026.0663.342.9
Global Entropy RegModel=Qwen3-4B-Base2026.0653.340.3
BaseModel=Qwen3-4B-Base2026.065038.4
PAECModel=Llama3.1-8B2026.066.71
GRPOModel=Llama3.1-8B2026.063.30.2
Global Entropy RegModel=Llama3.1-8B2026.063.30.5
BaseModel=Llama3.1-8B2026.0600.1