Exams / AIF-C01 / Prompt engineering / Question 2
AIF-C01 practice question: Prompt engineering (2 of 4)
Applications of Foundation Models A free question from our original AIF-C01 practice exam. Try it, then open the answer to see why each option is right or wrong.
An accounting firm finds that a model often gets multi-step calculation questions wrong when asked for the answer directly. Which prompting technique is MOST likely to improve accuracy on these questions?
Choose one answer.
- AA. Ask the model to reason step by step before giving the final answer
- BB. Ask the model to answer in as few words as possible, with no working
- CC. Remove all background context so the model is not distracted
- DD. Raise the temperature so the model tries more possible answers
Show the answer and explanation
Answer: A
- A. Right: Correct. Chain-of-thought prompting guides the model through intermediate reasoning, which improves accuracy on multi-step problems.
- B. Wrong: Forcing a very short answer removes room for reasoning and tends to make multi-step errors worse.
- C. Wrong: Removing the needed figures or context makes correct answers less likely.
- D. Wrong: More randomness makes answers less consistent. It does not improve step-by-step reasoning.
Explanation
Chain-of-thought prompting asks the model to work through a problem step by step before answering. It improves accuracy and makes the reasoning easier to check on multi-step tasks such as maths and analysis. The trade-off is more output tokens, which adds latency and cost, so it is not needed for simple tasks.
Study
Keep practising
Answer questions with a score at the end, or practise just this topic with the explanation after each one. No sign-up needed.