Fundamentals

Chain-of-Thought

Chain-of-thought (CoT) prompting instructs the model to explicitly show its reasoning process before providing the final answer. Instead of jumping to a conclusion, the model writes out intermediate steps. This simple technique dramatically improves performance on complex reasoning tasks, especially math and logic problems. CoT works because it forces the model to decompose problems into smaller steps, reducing errors. You can encourage CoT by saying 'Let's think step by step' or by showing examples of step-by-step reasoning in few-shot prompts.