How Retrieval-Augmented Generation (RAG) is Revolutionizing AI
How It Got Named ‘RAG’
Patrick Lewis, lead author of the 2020 paper that coined the term, apologized for the unflattering acronym that now describes a growing family of methods across hundreds of papers and dozens of commercial services he believes represent the future of generative AI.
So, What Is Retrieval-Augmented Generation (RAG)?
Retrieval-augmented generation (RAG) is a technique for enhancing the accuracy and reliability of generative AI models with facts fetched from external sources.
Combining Internal, External Resources
Lewis and colleagues developed retrieval-augmented generation to link generative AI services to external resources, especially ones rich in the latest technical details.
Building User Trust
Retrieval-augmented generation gives models sources they can cite, like footnotes in a research paper, so users can check any claims. This builds trust.
How People Are Using RAG
With retrieval-augmented generation, users can essentially have conversations with data repositories, opening up new kinds of experiences. This means the applications for RAG could be multiple times the number of available datasets.
Getting Started With Retrieval-Augmented Generation
To help users get started, NVIDIA developed an AI Blueprint for building virtual assistants. Organizations can use this reference architecture to quickly scale their customer service operations with generative AI and RAG, or get started building a new customer-centric solution.
The History of RAG
The roots of the technique go back at least to the early 1970s. That’s when researchers in information retrieval prototyped what they called question-answering systems, apps that use natural language processing (NLP) to access text, initially in narrow topics such as baseball.
Insights From a London Lab
The seminal 2020 paper arrived as Lewis was pursuing a doctorate in NLP at University College London and working for Meta at a new London AI lab. The team was searching for ways to pack more knowledge into an LLM’s parameters and using a benchmark it developed to measure its progress.
How Retrieval-Augmented Generation Works
At a high level, here’s how an NVIDIA technical brief describes the RAG process. When users ask an LLM a question, the AI model sends the query to another model that converts it into a numeric format so machines can read it. The numeric version of the query is sometimes called an embedding or a vector.
Keeping Sources Current
In the background, the embedding model continuously creates and updates machine-readable indices, sometimes called vector databases, for new and updated knowledge bases as they become available.
Conclusion
The future of generative AI lies in creatively chaining all sorts of LLMs and knowledge bases together to create new kinds of assistants that deliver authoritative results users can verify.
Frequently Asked Questions
Q: What is retrieval-augmented generation (RAG)?
A: RAG is a technique for enhancing the accuracy and reliability of generative AI models with facts fetched from external sources.
Q: How does RAG work?
A: RAG combines LLMs with embedding models and vector databases to retrieve relevant information from external sources and incorporate it into the LLM’s response.
Q: What are the benefits of RAG?
A: RAG builds trust by providing sources for the information it generates, reducing the possibility of hallucination, and enabling users to verify the accuracy of the results.
Q: How can I get started with RAG?
A: You can start by using NVIDIA’s AI Blueprint for building virtual assistants, which provides a reference architecture for scaling customer service operations with generative AI and RAG, or by exploring LangChain, an open-source library for chaining LLMs and knowledge bases together.
