RAG (Retrieval-Augmented Generation)

What is RAG (Retrieval-Augmented Generation)?

RAG is a technique that retrieves relevant data from an external source and provides it as context to the LLM, grounding the model's answers in facts and reducing hallucinations.

Where did the term "RAG (Retrieval-Augmented Generation)" come from?

Proposed by Facebook AI Research (FAIR) in 2020.

How is "RAG (Retrieval-Augmented Generation)" used today?

The standard architecture for building enterprise AI applications.

Related Terms

• Hallucination
• Vector Embeddings
• semantic-search