Retrieval-Augmented Generation
Retrieve relevant passages from a knowledge base, then have the model answer from them.
In plain language
RAG connects external knowledge to generation, suiting document QA, support knowledge bases and internal search. Core steps include chunking, embedding, vector retrieval, reranking and cited generation. RAG does not guarantee correctness; retrieval and source quality set the ceiling.
Example
A support bot retrieves three passages from the product manual, answers from them and cites sources.
Recommended tools
Related terms
Embedding
Converting text, images or other content into vectors whose similarity can be computed.
Fine-tuning
Continue training a model on task data to better match a style or task.
RAG vs Fine-tuning
RAG handles changing knowledge; fine-tuning handles behavior and style customization.
Vector Database
A database built to store vectors and retrieve them by similarity.