Basics of Prompt Engineering AIE Practice Questions (Page 3)

Questions 11–15

1. 11application · medium

   A customer support team wants to use an AI assistant to classify incoming emails into categories: Billing, Technical, or Other. They have a set of 10 labeled example emails for each category. Which prompt pattern should they use to maximize classification accuracy?

   • AZero-shot prompting, asking the model to classify each email without examples.
   • BChain-of-thought prompting, asking the model to reason step-by-step before classifying.
   • CInstruction-only prompting, giving a detailed description of each category without examples.
   • DFew-shot prompting, providing a few labeled examples in the prompt before asking the model to classify a new email.
   Reveal Solution Discussion Select an answer first
2. 12foundation · easy

   What is the first step in evaluating the effectiveness of a prompt?

   • ASelecting a different AI model to compare responses
   • BAssessing whether the model's response meets the desired criteria
   • CRewriting the prompt with different wording
   • DChanging the model's parameters to see if the output improves
   Reveal Solution Discussion Select an answer first
4. 13application · medium

   A teacher wants to use an AI assistant to generate quiz questions from a textbook chapter. The teacher has a few example questions from previous quizzes. Which prompt pattern would best help the AI generate questions in the same style?

   • AChain-of-thought prompting, asking the model to explain why each question is good.
   • BFew-shot prompting, including the example questions in the prompt.
   • CInstruction-only prompting, describing the desired style in words.
   • DZero-shot prompting, asking for questions without examples.
   Reveal Solution Discussion Select an answer first
5. 14expert · hard

   A financial analyst needs an AI to extract key metrics from quarterly reports. The analyst has a limited prompt budget (token limit) and needs to process many reports. They want consistent output format. Which approach balances accuracy and token efficiency?

   • AUse chain-of-thought prompting for every report.
   • BInclude the entire quarterly report in the prompt.
   • CUse a few-shot prompt with one example of a correctly extracted metric table.
   • DUse a zero-shot prompt with a detailed description of the metrics.
   Reveal Solution Discussion Select an answer first
6. 15expert · hard

   An AI-powered support chatbot frequently gives incorrect answers about product return policies. The team has tried adding more instructions, but accuracy remains low. They suspect the model lacks the latest policy details. Which approach is most effective within prompt engineering?

   • AAsk the model to answer only with 'I don't know' when uncertain.
   • BAdd a few-shot example of a correct return-policy answer.
   • CEmbed the full return policy text in the system prompt.
   • DUse chain-of-thought prompting to reason through the policy.
   Reveal Solution Discussion Select an answer first