Definition

A technique used with Large Language Models (LLMs) to enhance the accuracy and relevance of generated text by first retrieving pertinent information from an external knowledge base.

Why it matters (in Poovi’s context)

Significantly improves the factual grounding of LLM outputs, reduces ‘hallucinations’, and allows models to leverage up-to-date and domain-specific information beyond their initial training data.

Key properties or components

  • Retrieves external information
  • Augments LLM generation
  • Improves factual accuracy

Contradictions or debates

None.

Sources