Back to GlossaryApplications

Chain of Thought

Definition

A prompting technique that encourages AI models to break down complex reasoning into intermediate steps before arriving at a final answer, significantly improving accuracy on logical and mathematical tasks.

Chain of Thought (CoT) prompting was introduced by Google researchers in 2022 and demonstrated that simply asking a model to "think step by step" dramatically improves performance on math, logic, and multi-step reasoning problems. Instead of jumping directly to an answer, the model shows its work, making errors easier to identify and correct. CoT can be elicited through few-shot examples showing step-by-step reasoning or through zero-shot instructions like "Let's think through this step by step." Advanced variants include Tree of Thought (exploring multiple reasoning paths), self-consistency (sampling multiple CoT paths and taking the majority answer), and extended thinking modes built into models like Claude. CoT reasoning has become a standard capability in modern LLMs.

Companies in Applications

View Applications companies →