What is RAG (Retrieval-Augmented Generation)?
TL;DR
A technique that improves AI response accuracy by retrieving relevant information from external databases before generating answers.
RAG (Retrieval-Augmented Generation): Definition & Explanation
RAG (Retrieval-Augmented Generation) is a technique designed to improve the accuracy of LLM responses. When a user asks a question, the system first retrieves relevant information from external databases or documents, then includes that information in the LLM's input to generate a more accurate answer. This enables the AI to provide precise responses based on up-to-date information or private company data that the LLM was not trained on. RAG helps reduce hallucinations (fabricated information) and makes it possible to cite specific sources for each answer. It is widely used in enterprise chatbots and knowledge base search systems.