Techniques
Retrieval-Augmented Generation (RAG)
Also known as: retrieval augmented generation
In one line
A technique where the AI looks things up in your documents before answering.
What does Retrieval-Augmented Generation (RAG) mean?
RAG combines a search step (retrieve relevant chunks from a vector database) with a generation step (feed those chunks into an LLM). It lets AI answer from private data without retraining.
A real-world example
A support bot that searches your help centre for the closest article, then rewrites it into a friendly answer.