Retrieval-augmented generation
Searching a collection of documents for relevant passages and giving them to the model along with the question.
Also: RAG
Retrieval-augmented generation was described by Patrick Lewis and colleagues at Facebook AI Research in 2020. A question is turned into an embedding, the closest passages are found in a vector database or search index, and they are placed in the prompt so the model answers from them. RAG lets a model use private or recent information without retraining and makes answers easier to check. Results depend heavily on retrieval quality: if the right passage is not found, the model may answer from memory or guess.