AWS Certified Generative AI – Professional (AIP-C01) — Question 23

A company uses Amazon Bedrock to generate technical content for customers. The company has recently experienced a surge in hallucination outputs when the company's model generates summaries of long technical documents. The model outputs include inaccurate or fabricated details. The company's current solution uses a large foundation model (FM) with a basic one-shot prompt that includes the full document in a single input.
The company needs a solution that will reduce hallucinations and meet factual accuracy goals. The solution must process more than 1,000 documents each hour and deliver summaries within 3 seconds for each document.
Which combination of solutions will meet these requirements? (Choose two.)

Answer options

• A. Implement zero-shot chain-of-thought (CoT) instructions that require step-by-step reasoning with explicit fact verification before the model generates each summary.
• B. Use Retrieval Augmented Generation (RAG) with an Amazon Bedrock knowledge base. Apply semantic chunking and tuned embeddings to ground summaries in source content.
• C. Configure Amazon Bedrock guardrails to block any generated output that matches patterns that are associated with hallucinated content.
• D. Increase the temperature parameter in Amazon Bedrock.
• E. Prompt the Amazon Bedrock model to summarize each full document in one pass.

Correct answer: A, B

Explanation

Option A is correct as it encourages detailed reasoning and fact verification, which can significantly reduce hallucinations. Option B is also correct because it employs RAG to ensure the generated summaries are grounded in accurate source material. Options C and D do not directly address the issue of hallucinations effectively, while option E continues the existing one-shot approach that has proven problematic.