Guide the model to reason step-by-step before answering — dramatically improving accuracy on math, logic, and multi-step tasks.
Five passes over the same idea, each from a different angle. Do them in order, or jump to whichever you need.
Chain-of-thought (CoT) prompting asks the model to show its reasoning process before giving a final answer. This simple technique improves accuracy on arithmetic, symbolic reasoning, commonsense inference, and complex multi-step tasks. Variants include zero-shot CoT ("Let's think step by step"), manual CoT (with example reasoning chains), and self-consistency (sample multiple chains, pick the majority answer).
