List your company

Math Reasoning on BRUMO 2025 (pass@1, pass@4)

4.79Pass@1GRPO
3.493.82754.1654.5025Oct 8, 2025

Evaluation results

MethodLinks
GRPOModel=Llama-3.2-3B2025.104.797.82
XRPOModel=Llama-3.2-3B2025.103.549.48