Retrieval-Augmented Generation (RAG)

Entity Type: Glossary ID: retrieval-augmented-generation

Definition: A framework that enhances language models by combining parametric knowledge stored in model weights with non-parametric knowledge retrieved from external databases or knowledge bases. RAG systems first retrieve relevant documents or passages based on input queries, then use this retrieved information to augment the generation process, enabling models to access up-to-date and domain-specific information beyond their training data.

Related Terms: - information-retrieval - knowledge-bases - language-models - vector-databases - embedding-models

Source Urls: - https://arxiv.org/abs/2005.11401 - https://huggingface.co/docs/transformers/model_doc/rag - https://python.langchain.com/docs/modules/data_connection/

Tags: - text-generation - search - knowledge-bases - language-models

Status: active

Version: 1.0.0

Created At: 2025-09-10

Last Updated: 2025-09-10