Question

23,764 Learners found this answer helpful

what is retrieval augmented generation?

Explanation:

Define the core concept

Retrieval-Augmented Generation (RAG) optimizes large language model outputs. It references authoritative external knowledge bases.

Explain the retrieval step

The system queries external sources first. It gathers relevant documents matching the user's prompt.

Explain the generation step

The model receives retrieved documents. It synthesizes a factual, context-grounded response.

Highlight key benefits

RAG reduces model hallucinations. It allows real-time updates without retraining.

Provide a practical analogy

Think of an open-book exam. The model searches the textbook before answering.

Answer:

Retrieval-Augmented Generation (RAG) is an artificial intelligence framework that improves the accuracy and reliability of Large Language Models (LLMs) by fetching facts from an external, authoritative knowledge base before generating a response.

Instead of relying solely on its static training data, a RAG system works in two main phases:

1. Retrieval: When a user asks a question, the system searches an external database or document collection to find information relevant to the query.
2. Generation: The system appends this retrieved context to the user's original prompt and passes it to the LLM, which synthesizes a highly accurate, up-to-date, and context-grounded answer.