Skip to course content
Free RAG fundamentals course

Introduction to RAG and Knowledge Assistants

Module 06 Knowledge Check

5 questions. Pass mark 4 out of 5. Answer every question before checking the answer key below, then retry after reading the feedback.

1. When the corpus does not contain the answer, the correct behaviour is…

2. Why is 'no answer found' hard to achieve in practice?

3. A useful mechanism for refusal is…

4. A good 'I don't know' response should also…

5. Measuring refusal quality requires test questions that…

---

Answer Key and Explanations

Check these only after attempting every question.

1. B - Say the information is not available in the sources and stop

A well-designed assistant must be able to say it does not know; this is a feature, not a failure.

**2. B - Retrieval almost always returns *something*, and the model tends to use it**

Vector search returns nearest neighbours even when nothing is genuinely relevant - the top result is not necessarily a good one.

3. B - A relevance threshold or check that can return zero usable passages

You need an explicit path where retrieval yields nothing acceptable, rather than always passing the top-k forward.

4. B - Say what was searched and suggest the next step or human route

An actionable refusal keeps the user moving and shows the boundary of the system's knowledge.

5. B - Include questions deliberately outside the corpus

If every eval question is answerable, you never test the most dangerous behaviour: confident invention.

Practical Check

Apply this module to your own work: complete the module activity for *Missing Answers and Refusal Behaviour*, then write one sentence naming what your result shows and one naming what it does not.

Strong Answer Pattern

A strong answer names the task, the evidence used, the check performed, and the remaining limitation. It avoids "proved", "guaranteed", or "always" unless the evidence genuinely supports it.