List your company

Language Understanding on MMLU PRO (ACC. and LEN.)

17.58AccuracyGRPO
17.403217.449117.49517.5409Mar 19, 2026

Evaluation results

MethodLinks
GRPOModel=Qwen3-14b2026.0317.581,817
DDPOModel=Qwen3-14b, theta=0.22026.0317.411,548