AI Giants Go Nuclear, A Tech Bromance Turns Turbulent, Mistral Sharpens the Edge, and more...

Suppressing Pink Elephants with Direct Principle Feedback

Direct Preference Optimization: Your Language Model is Secretly a Reward Model | DPO paper explained

The third New England RLHF Hackers Hackathon

Scientists Are Researching a Device That Can Induce Lucid Dreams on Demand

Yann LeCun on X

Human Feedback is not Gold Standard

OpenELM/OpenELM_Paper.pdf at paper · CarperAI/OpenELM · GitHub

SelFee: Iterative Self-Revising LLM Empowered by Self-Feedback Generation

Quicker feedback on new content

Addressing criticism, OpenAI will no longer use customer data to train its models by default

The Flan Collection: Advancing open source methods for instruction tuning

Charlie George on Twitter