RAG (Retrieval-Augmented Generation)

RAG connects LLMs to enterprise data via vector search so responses are grounded in proprietary knowledge bases.