liveBook · Manning

Retrieval-augmented generation

As explained in Rearchitecting LLMs

Retrieval-augmented generation is a technique that retrieves external information at generation time and provides it to a language model as context.

Retrieval-augmented generation (RAG) gives a language model access to external information at generation time. It is useful for integrating up-to-date and private knowledge, but it does not transform the underlying model. section 1.1.2

RAG keeps information processing generic, so it may not provide differentiated performance by itself. Its value increases when the model is adapted to retrieve, interpret, and use external information correctly within the pipeline. section 1.1.2

How Rearchitecting LLMs covers it

How deep it goes
Defines the technique and its trade-offs, but stops short of implementation details, retrieval design, and evaluation procedures.
Examples
Uses private organizational data and up-to-date information as the concrete knowledge sources.
What you can do after
Apply it via a RAG pipeline to give a language model access to external information.
How it connects
Connects retrieval-augmented generation with prompt engineering, fine-tuning, and model adaptation.

Also explained in

Compare how 46 books explain retrieval-augmented generation →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Rearchitecting LLMs. The section links point to the parts of the book it draws on.