Fusing Pre-Trained Language Models With Multimodal Prompts Through Reinforcement Learning | Read Paper on Bytez