Chain of Thought
In plain terms
Prompting a model to reason step by step.
Going deeper
Chain-of-thought prompting elicits intermediate reasoning steps from a model, improving accuracy on complex tasks. Showing the work helps it think.
Under the hood
Improves multi-step reasoning; related techniques include self-consistency and tree-of-thought.
Examples
Math/logic problems, complex reasoning, prompt engineering, LLM accuracy.
Think of it like...
Like asking a student to show their working rather than just the final answer.