AI-900 Practice Question: Describe features of generative AI workloads on Azure
What is 'few-shot prompting' and how does it improve model outputs?
⚠ Common exam trap
Many exam-takers confuse 'few-shot' with 'fewer training data' or 'shorter responses,' when the term specifically refers to the number of examples included in the prompt to guide the model's output.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Including a small number of input-output examples in the prompt to demonstrate the desired task format
Few-shot prompting improves model outputs by providing a small number of input-output examples directly in the prompt, which helps the model understand the desired task format, style, or reasoning pattern without requiring any fine-tuning or retraining. This technique leverages the model's in-context learning ability to generalize from the given examples and produce more accurate, consistent responses.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Training a model with very few labelled examples using transfer learning
Why it's wrong here
Few-shot prompting is a pure inference-time technique; it does not update model weights or require gradient-based training. Transfer learning with very few labeled examples typically refers to fine-tuning a pretrained model on a small dataset, which still involves a training loop and adjusting parameters. The confusion here is equating in-context demonstrations with parameter-level learning, which are fundamentally different mechanisms.
- ✓
Including a small number of input-output examples in the prompt to demonstrate the desired task format
Why this is correct
Few-shot prompting means supplying a small number of complete input-output pairs directly in the prompt so the language model can infer the task pattern by conditioning on those examples. No weights are modified, and no optimization step occurs — the model simply uses the provided demonstrations as context to generate a matching output for a new input. This is distinct from transfer learning because it relies on the model's existing emergent abilities rather than further training on labelled data.
- ✗
Generating a short (few-shot) response rather than a detailed answer
Why it's wrong here
The term 'few-shot' in prompt engineering refers to the number of demonstrations (examples) placed in the prompt, not the length of the generated answer. Generating a short response is a matter of controlling verbosity via instructions or token limits, which is an unrelated concern. Therefore, interpreting 'few-shot' as response brevity misunderstands the core concept, which is about providing a few examples to guide the model's output format and behaviour.
- ✗
Running the model for only a few seconds to save compute costs
Why it's wrong here
Few-shot prompting has nothing to do with runtime duration; it is a prompt-engineering strategy aimed at improving output quality through example demonstrations. Running a model for only a few seconds is a compute-cost or latency optimization, which is orthogonal to how the prompt is structured. In fact, few-shot prompting often increases the number of input tokens, potentially increasing inference cost rather than decreasing it.
Go deeper
Related to this question
About these practice questions
One of 985 original AI-900 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI-900 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI-900 exam.