Retrieval-Augmented Generation (RAG)

A technique where an LLM fetches relevant documents before generating an answer.