← All 25 termsAI proof
RAG (Retrieval-Augmented Generation)
AIA pattern where a language model is given retrieved context (chunks from a vector store) before generating its answer. Reduces hallucination on factual queries.
Full definition
Retrieval-Augmented Generation (RAG) pairs a language model with a vector store: at inference time, the query is embedded, the top-k most similar chunks are retrieved, and those chunks are inserted into the model's context window as grounding. RAG reduces hallucination on factual queries because the model has access to the source material rather than relying on its parametric memory. On this site, the qfin-rag-harness is a reference RAG stack for financial-document QA with an eval harness that scores retrieval recall and answer faithfulness.
Related terms
Each link opens the long-form definition of a closely related term.
See this term in action.
The /proof page maps every term to the gate or artifact that implements it on this site.