Combining document retrieval with LLM generation for grounded answers.
Retrieval-Augmented Generation (RAG) is an architecture that pairs a language model with a retrieval system, so the model can pull relevant documents at query time and generate an answer grounded in that external information. RAG reduces hallucinations and lets AI systems cite up-to-date or proprietary sources they were not trained on. Answer engines like Perplexity and AI search features rely heavily on retrieval to decide which sources to quote.