Vision-Language Models are Zero-Shot Reward Models for Reinforcement Learning Paper • 2310.12921 • Published Oct 19, 2023 • 19