DOLA: A SIMPLE WAY TO MAKE LANGUAGE MODELS STICK TO THE FACTS

Summary from the TLDR AI newsletter

Researchers have introduced a new method called DoLa that helps large language models (like the ones that chat with you or write articles) stop making stuff up and stick to the facts. It uses a smart way of comparing different layers inside the model to filter out false information.