Retrieval-Augmented Generation CT-GENAI Practice Questions (Page 6)

Questions 26–28

1. 26application · medium

   A team evaluates their RAG system for test case generation. They measure retrieval accuracy by checking if the top 5 retrieved documents contain the relevant test case. What metric are they using?

   • APrecision@5
   • BF1 score
   • CRecall@5
   • DMean Reciprocal Rank (MRR)
   Reveal Solution Discussion Select an answer first
2. 27foundation · easy

   What is the primary purpose of chunking documents before embedding them for retrieval?

   • ATo increase the size of the vector database
   • BTo make the LLM generate longer responses
   • CTo encrypt the documents for security
   • DTo break documents into smaller pieces that can be embedded and retrieved effectively
   Reveal Solution Discussion Select an answer first
3. 28expert · hard

   A team evaluates their RAG system for test analysis. They measure retrieval precision and recall, but the end-to-end performance is still poor because the LLM generates irrelevant analysis. What should they do next?

   • AEvaluate generation relevance and adjust the prompt to better use the retrieved context.
   • BImprove retrieval by using a different embedding model.
   • CReduce the number of retrieved documents.
   • DIncrease the number of retrieved documents.
   Reveal Solution Discussion Select an answer first