Chain of Thought, Teaching Models to Think Out Loud

Paper: Chain-of-Thought Prompting Elicits Reasoning in Large Language Models (2022)

This paper introduced one of the simplest and most powerful tricks in all of AI, and the wonderful part is that it requires no new training at all. You just change how you ask. The idea is to get the model to show its work before giving an answer, exactly like a math teacher asking you to write out the steps. That single change unlocks a dramatic jump in reasoning ability.