Retrieval-augmented generation

Retrieval-augmented generation (RAG) is a technique that enables <a href="/facts/Large_language_model/WnogWVJY">large language models</a> (LLMs) to retrieve and incorporate new information. With RAG, LLMs do not respond to user queries until they refer to a specified set of documents. These documents supplement information from the LLM's pre-existing <a href="/facts/Training_data/V3mMQb2G">training data</a>. This allows LLMs to use domain-specific and/or updated information that is not available in the training data. For example, this helps LLM-based <a href="/facts/Chatbot/kASkMjs7">chatbots</a> access internal company data or generate responses based on authoritative sources.
RAG improves large language models (LLMs) by incorporating <a href="/facts/Information_retrieval/MFSmpeNw">information retrieval</a> before generating responses. Unlike traditional LLMs that rely on static training data, RAG pulls relevant text from databases, uploaded documents, or web sources. According to <a href="/facts/Ars_Technica/zvnX7pnh">Ars Technica</a>, "RAG is a way of improving LLM performance, in essence by blending the LLM process with a web search or other document look-up process to help LLMs stick to the facts." This method helps reduce <a href="/facts/AI_hallucinations/gwBoRg6p">AI hallucinations</a>, which have caused chatbots to describe policies that don't exist, or recommend nonexistent legal cases to lawyers that are looking for citations to support their arguments.
RAG also reduces the need to retrain LLMs with new data, saving on computational and financial costs. Beyond efficiency gains, RAG also allows LLMs to include sources in their responses, so users can verify the cited sources. This provides greater transparency, as users can cross-check retrieved content to ensure accuracy and relevance.
The term RAG was first introduced in a 2020 research paper from <a href="/facts/Meta_(Company)/ylApt16y">Meta</a>.

Retrieval-augmented generation open-in-new

Retrieval-augmented generation