nerdexam
Databricks

GENERATIVE-AI-ENGINEER-ASSOCIATE · Question #85

Which of the following are key considerations when identifying source documents for a RAG application? (Choose two.)

The correct answer is A. Information density D. Document relevance to the task. When selecting source documents for RAG, the content quality and relevance to the task matter far more than physical document properties like size or format.

Building Retrieval-Augmented Generation (RAG) Applications

Question

Which of the following are key considerations when identifying source documents for a RAG application? (Choose two.)

Options

  • AInformation density
  • BDocument size
  • CDocument format
  • DDocument relevance to the task

How the community answered

(39 responses)
  • A
    87% (34)
  • B
    8% (3)
  • C
    5% (2)

Why each option

When selecting source documents for RAG, the content quality and relevance to the task matter far more than physical document properties like size or format.

AInformation densityCorrect

Information density ensures that retrieved chunks contain meaningful, actionable content rather than noise or filler text, which directly improves the signal-to-noise ratio of context passed to the LLM.

BDocument size

Document size is a preprocessing and chunking concern, not a source selection criterion - large documents can be chunked and small documents can still be high quality.

CDocument format

Document format affects the parsing and ingestion pipeline but does not determine whether a source is appropriate for the RAG application's knowledge base.

DDocument relevance to the taskCorrect

Document relevance to the task is the foundational criterion for RAG source selection - including off-topic documents pollutes the vector index and causes the retriever to surface irrelevant context, degrading response quality.

Concept tested: RAG source document selection criteria

Source: https://docs.databricks.com/en/generative-ai/retrieval-augmented-generation.html

Topics

#RAG#Data Preparation#Information Retrieval#Source Document Selection

Community Discussion

No community discussion yet for this question.

Full GENERATIVE-AI-ENGINEER-ASSOCIATE Practice