JoMoHa: Joint Model–Harness Optimization with Reinforcement Learningwith Xudong Wu, Guangchen Lan, Mingqi Yuan1 versionW_yuqqd4bjv1 · current