Arash Ahmadian on Rethinking RLHF

od TalkRL: The Reinforcement Learning Podcast