Retrieval-Augmented Generation
An AI architecture that retrieves relevant documents from external sources and uses them as context when generating responses.
Retrieval-Augmented Generation (RAG) is an architecture used by AI answer engines like ChatGPT and Perplexity. When a user submits a query, the system converts it into a vector representation, searches a knowledge database or live search index for relevant documents, and then augments the language model's prompt with those retrieved passages before generating a response. RAG enables AI systems to cite current, authoritative sources rather than relying solely on training data.
Also known as
RAG, retrieval augmented generation, retrieval-augmented generation