RAG

In AI, "RAG" stands for "Retrieval-Augmented Generation," which is a technique that enhances the capabilities of a large language model (LLM) by allowing it to access and reference external knowledge bases, like a company's internal database, before generating a response, resulting in more accurate and contextually relevant outputs compared to relying solely on its trained data.