Retrieval-Augmented Generation

Technique that lets large language models retrieve and incorporate fresh information from external data sources before answering.

Specifications & Metrics

type

Agents & Systems

symbol

RAG

Core Principles & Analysis

• Retrieval-augmented generation (RAG) is a technique that enables large language models to retrieve and incorporate new information from external data sources.
• It combines a retriever over external knowledge with the generator.
• Agentic RAG extends it with agent-driven, multi-step retrieval.

Sources

• Wikipedia: Retrieval-augmented generation