Retrieval-Augmented Generation

The Retrieval-Augmented Generation Process

This page is currently in draft

What is Retrieval-Augmented Generation (RAG) ?

Retrieval-Augmented Generation (RAG) is a method for improving the accuracy and reliability of generative AI models by applying facts fetched from authoratitive sources.

The key difference betweem the general AI of Large Language Models is that data retrival is augmented (or supplemented) by local or contextual data and rules.

The term RAG was coined in a 2020 paper entitled - "Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks"