AIF-C01 Fundamentals of Generative AI Practice Question
A company deployed a question-answering system using Amazon Bedrock with a knowledge base (RAG). Users report that the model often hallucinates facts not in the knowledge base. What is the most effective way to reduce hallucinations?
⚠ Common exam trap
A common misconception is that hallucinations are primarily a model training issue (fine-tuning or context length) rather than a retrieval quality issue in RAG systems, leading candidates to overlook the critical role of the retriever in grounding responses.
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
✓
Improve the relevance of retrieved documents by refining the retrieval strategy
Hallucinations in RAG systems often stem from the model receiving irrelevant or low-quality retrieved documents, which forces it to rely on its parametric knowledge rather than the provided context. By refining the retrieval strategy—such as improving embedding quality, adjusting chunk overlap, or using hybrid search—the system ensures the foundation model has the most relevant information to ground its answers, directly reducing the likelihood of fabricating facts.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Reduce the maximum context length to limit model input
Why it's wrong here
Truncating the context removes the retrieved evidence the model needs, so it answers from parametric memory and hallucinates more. It is tempting because limiting input genuinely controls token cost and latency, but the stem's requirement is factual grounding, which demands retaining retrieved passages rather than discarding them.
- ✗
Fine-tune the foundation model on a large general corpus
Why it's wrong here
Fine-tuning on a general corpus teaches the model broader world knowledge, which increases confident fabrication rather than grounding answers in retrieved documents. It is tempting because fine-tuning is genuinely used to adapt a model's style or domain vocabulary, but it does not constrain generation to the knowledge base at inference time.
- ✓
Improve the relevance of retrieved documents by refining the retrieval strategy
Why this is correct
Refining retrieval directly targets the RAG constraint: hallucinations arise when retrieved context lacks the facts needed, so the model fabricates answers. Improving retrieval relevance—through better chunking, embeddings or hybrid search—ensures grounding documents actually contain the answer, reducing fabrication without changing the foundation model itself.
- ✗
Increase the chunk size of documents in the knowledge base
Why it's wrong here
Larger chunks dilute retrieval precision, so irrelevant or partial passages reach the prompt and the model fills gaps with invented content. It is tempting because bigger chunks are genuinely useful when answers span long passages and context windows are generous, but the requirement here is tighter grounding, which smaller, more precise chunks serve.
Go deeper
Related to this question
About these practice questions
This AIF-C01 question is part of Courseiva's 862-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AIF-C01 practice question is part of Courseiva's free Amazon Web Services 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 AIF-C01 exam.