DeepMind: the existence proof for RL at scale, by Nathan Lambert

Import AI 333: Synthetic data makes models stupid; chatGPT eats MTurk. Inflection shows off a large language model

Reward is not enough - by Nathan Lambert - Interconnects

29th Annual Computational Neuroscience Meeting: CNS*2020, BMC Neuroscience

Arun Rao (@rao_hacker_one) / X

Deep RL Case Study: Model-based Planning, by Nathan Lambert

Deep learning is not the key to unlocking the Singularity, by Nathan Lambert

awesome-transformer-nlp/README.md at master · cedrickchee/awesome-transformer-nlp · GitHub

Frontiers Learning and Animal Movement

arxiv-sanity

Open Problems and Fundamental Limitations of Reinforcement Learning From Human Feedback, PDF, Artificial Intelligence

All stories published by Towards Data Science on April 26, 2020

Nathan Lambert on X: New paper! We outline my argument as to why more transparency and open-source action around reward models is so crucial to the development of RLHF. Entangled Preferences: The

Jim Fan on LinkedIn: Human creations are sometimes too advanced for GPT-4V to appreciate. 🤣…