Progress Reward Model for Reinforcement Learning via Large Language Models | Read Paper on Bytez