Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
Source
Patrick Lewis et al.
Author
Patrick Lewis et al.
Published
Terms in this piece · Glossary
RAG — Retrieval-augmented generation — fetching relevant documents first and pasting them into the model's context so it answers from your data instead of memory.
Why it matters
The original formulation of what the industry now just calls RAGRetrieval-augmented generation — fetching relevant documents first and pasting them into the model's context so it answers from your data instead of memory.Full definition →: pair a generator with a dense retrieval index so knowledge lives in a store you can update rather than in weights you must retrain. Worth reading to see how much of what people improvise was specified here.
articleBM25 Wins at Scale: A Scaling Study of Retrieval-Augmented Generation ParadigmsPengyu Wang, Benfeng Xu, Shaohan Wang, Xin Zeng, Huarui Wu, Lei Zhang, Licheng Zhang
articleWhen Knowledge Changes: Metamorphic Testing of RAG Systems with MutationsJinhan Kim, Samuele Pasini, Paolo Tonella
articleUniversal Pathologies, Conditional Consequences: A Triple-Robustness Analysis of RAG for Multi-Hop TraceabilityMeftun Akarsu, Burak Ozdemir