Retrieval-Augmented Generation (RAG)

Retrieval-augmented generation combines information retrieval with generation. It often searches a vector database and supplies evidence to a large language model, improving grounding without retraining the model for every knowledge change.

Japanese terms

1. Retrieval-augmented generation — 検索拡張生成けんさくかくちょうせいせい: A generation method that retrieves external information and supplies it to a generative model as context.
2. Information retrieval — 情報検索じょうほうけんさく: Finding relevant documents or records in response to an information need.
3. Vector database — ベクトルデータベース: A database designed to store embeddings and retrieve nearby vectors efficiently.
4. Grounding — グラウンディング: Connecting a model’s output to supplied evidence, data, or observable context.

Related topics

Embeddings commonly power semantic retrieval for RAG systems.