Chain of thought

Also: CoT, step-by-step reasoning

Having a model write out intermediate steps before its answer, which gives it more computation and a place to keep partial results. It helps most on arithmetic, multi-step logic and anything with a chain of constraints.

What people get wrong

The written reasoning is not a faithful record of how the answer was reached. It can be plausible, wrong, and still lead to the right answer — so it is a performance aid, not an explanation you can audit.

A definition is not an understanding. This is taught properly in Reasoning, and improving a prompt honestly, part of Prompting.

Learn this properly

Terms from the same topic