n. / Retrieval
RAG (retrieval-augmented generation)
Definition
Retrieval-augmented generation is a pattern where your system first looks up relevant documents and then gives them to the model along with the question, so the answer is based on that material instead of the model's memory alone.
In more detail
It lets a model answer from private or recent data without retraining. Quality depends on two parts: whether retrieval found the right passages, and whether the model used them faithfully. When an answer is wrong, a trace tells you which part failed, because it shows the retrieved text next to the final reply.
Example
A user asks about the refund window. The system retrieves three paragraphs from the policy handbook and the model answers "30 days", citing the second paragraph.
Related terms
