Exam-style
A subscription business uses a foundation model to work out refund amounts. The policy is tiered: the rate depends on how long the subscriber has been on the plan, an unused part-month is refunded pro rata, and the refund is capped at one month's fee. The prompt returns a single figure, and about one refund in five is wrong, usually because a step was skipped or the cap was applied before the pro-rata calculation instead of after. The team has already put eight worked examples in the prompt, each showing a subscriber, a policy tier and the finished figure, and the error rate has not moved. What should they change?
Reveal the answer
B. Ask the model to work through the policy steps in order and show the intermediate figures before the final amount
This is a reasoning failure, and examples do not fix reasoning. A few-shot prompt supplies finished input-output pairs, which pin down a format or an unusual category boundary. Single-shot supplies one pair, and zero-shot supplies instructions and no pairs at all. Eight subscriber-in, figure-out pairs have already shown what a finished answer looks like, and none of them show the working. Chain-of-thought prompting asks for the steps first: identify the tier, compute the rate, work out the unused days and the pro-rata amount, add them, apply the cap, state the result. The order of operations that keeps going wrong is then written down. Temperature adjusts randomness in token selection, and a refund calculation needs none. A prompt template is a fixed skeleton with slots for the subscriber and the plan, so it standardises how a prompt is assembled, not how the figure is worked out. Chain-of-thought generates the working as text, so every call returns more output tokens and takes longer. Keep the steps in a labelled section the application strips before the figure reaches a customer.
Q. A prompt turns a tiered refund policy into a figure and gets one in five wrong, usually by skipping a step or applying the cap in the wrong place. Eight worked examples did not help. What next?
A. Chain-of-thought. Ask for the tier, the rate, the pro-rata amount and the cap check as numbered steps, and the final figure last.
Why? Examples and reasoning fix different faults. Few-shot examples show what a finished answer looks like, and single-shot does the same with one pair. They settle format and awkward edge cases. Zero-shot gives instructions and nothing to copy. None of them show the working. Eight pairs demonstrate the shape of a refund figure and nothing about how to reach it. Chain-of-thought puts the intermediate steps in the output, so the order of operations is written down instead of skipped. Every call then generates more output tokens and takes longer to reach the figure. Raising temperature adds randomness to a calculation that needs none, and prompt templates standardise how the prompt is assembled and reused, not how the figure is worked out.