Skip to slide
12 / 53

CH 01 · Chain of Thought, Teaching Models to Think Out Loud · 07/08

Read as one page

Why this paper mattered

Chain of thought reframed what a prompt is for. A prompt is not just a question; it can shape how the model thinks. It showed that a huge amount of latent reasoning ability was already sitting inside large models, waiting to be unlocked by simply giving them room to work. Almost every later advance in reasoning, including the ReAct agents and the DeepSeek-R1 reasoning model in this folder, builds directly on this foundation of getting the model to produce intermediate steps.

Finished this idea?

Your learning progress stays in this browser.