Databricks Certified Generative AI Engineer Associate — Question 89

A team uses Mosaic AI Vector Search to retrieve documents for their Retrieval-Augmented Generation (RAG) pipeline. The search query returns five relevant documents, and the first three are added to the prompt as context. Performance evaluation with Agent Evaluation shows that some lower-ranked retrieved documents have higher context relevancy scores than higher-ranked documents.

Which option should the team consider to optimize this workflow?

Answer options

• A. Use a reranker to order the documents based on the relevance scores.
• B. Modify the prompt to instruct the LLM to order the documents based on the relevance scores.
• C. Use a different embedding model for computing document embeddings.
• D. Increase the number of documents added to the prompt to improve context relevance.

Correct answer: A

Explanation

The correct answer is A because using a reranker allows the team to reorder the documents based on their relevancy scores, ensuring the most relevant ones are prioritized in the prompt. Options B and D do not directly address the ranking issue, while option C, though potentially helpful, does not guarantee improved relevance ranking for the current context.