Skip to main content

RAG (Retrieval-Augmented Generation)

A method in natural language processing that combines traditional large language models (LLMs) with external information retrieval systems. In RAG, the model generates responses by first retrieving relevant information from a database or document collection and then using this information to inform or augment its output. This approach enhances the model's ability to provide accurate and up-to-date responses, especially when dealing with specific domains or the latest data. RAG systems are particularly useful for applications where the model needs to access information beyond its training data.

RAG data can be retrieved from vector- or relational databases, API's, documents,... and added to the context along with possible user- and system prompts

Definition generated by AI (Poolnoodle-Deeptought).