GLOSSARY
Retrieval-Augmented Generation (RAG)
A technique where a model first searches relevant documents (or the web) and then answers based on what it found, citing sources.
RAG combines a model's language ability with a search step: before answering, the system retrieves relevant passages from your documents or the live web and feeds them into the context window. The model then answers grounded in that material, usually with citations.
This is how NotebookLM answers only from your sources and Perplexity cites the web. RAG drastically reduces hallucination on factual questions and lets ordinary tools answer from knowledge that was never in their training data — your company wiki, this week's news, today's inventory.