← All artifacts · All patterns · Retrieval & RAG

Naive RAG

aka Retrieval-Augmented Generation, Top-K Retrieve-and-Stuff

category: retrieval · status: mature

Intent

Condition the generator on top-k chunks retrieved from an external dense index so knowledge lives outside parameters.

Context

The agent needs information that lives in a corpus too large to fit in context, and that may change without retraining.

Problem

Parametric LMs hallucinate, cannot cite, and cannot be updated without retraining; query-time external knowledge is needed.

Forces

• Chunk size trades context loss for retrieval recall.
• Embedding choice constrains retrieval quality.
• Single-shot retrieval misses multi-hop questions.

Solution

Chunk the corpus. Embed each chunk with a dense encoder. At query time, embed the query, retrieve top-k by similarity, prepend chunks to the prompt, generate. The simplest production RAG pipeline.

Constrains

Consequences

Benefits

• Knowledge updates without retraining.
• Citations become possible.

Liabilities

• Chunk boundaries destroy context.
• Top-k retrieval is recall-oriented; precision suffers without reranking.
• No iterative retrieval; multi-hop fails.

Known Uses

• LangChain / LlamaIndex default RAG
• Most enterprise document-QA deployments

Related Patterns

• generalises → hyde
• composes-with → cross-encoder-reranking
• generalises → contextual-retrieval
• alternative-to → graphrag
• specialises → agentic-rag
• conflicts-with → naive-rag-first — Naive RAG is fine; treating it as the only answer is the anti-pattern.
• composes-with → chain-of-verification
• generalises → vector-memory
• complements → citation-streaming
• generalises → raft
• generalises → hybrid-search
• alternative-to → hallucinated-citations
• used-by → app-exploration-phase

References

• Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks — Lewis, Perez, Piktus, Petroni, Karpukhin, Goyal, Küttler, Lewis, Yih, Rocktäschel, Riedel, Kiela (2020) paper