Retrieval-Augmented Generation (RAG)
Definition
RAG is an architecture that retrieves relevant documents at query time and conditions a language model’s answer on that retrieved context.
Full definition
Instead of relying only on parameters learned in training, RAG systems search a corpus (web, vector database, or docs), then generate an answer grounded in those passages. Many answer engines use variants of this pattern.
Site owners effectively compete to be retrieved: clear chunkable content, accurate titles, and crawl access matter. See Grounding and Answer Engine.