Yup! Actions > Words. Corollary: Reinforcement Learning > Large Language Models??

→ View original post on X — @oriolvinyalsml