LATENT REFERENCES / TAG1
rag retrieval augmented generation
Original title: rag retrieval augmented generation
This reference note belongs to Tag1 in Latent References, an archive curated by Keigo Yoshida. Its archive region is TouchDesigner. The note preserves its source text and links so that readers can trace the material behind the 3D map.
- Collection
- Tag1
- Archive region
- TouchDesigner
Archived reference note
English translation of the archived note. JP shows the original text. Source links and literal code are retained; the translation does not update or independently verify the source claims.
RAG (Retrieval-Augmented Generation) is a technique in which a large language model (LLM), before generating an answer, retrieves relevant information from external knowledge sources such as internal company documents and databases and constructs its answer based on that information. It enables accurate answers based on up-to-date and specialist information without retraining the model, and is strong in reducing hallucinations (falsehoods) and identifying information sources.
RAG（Retrieval-Augmented Generation、検索拡張生成）は、大規模言語モデル（LLM）が回答を生成する前に、社内文書やデータベースなどの外部知識源から関連情報を検索し、その情報を基に回答を構成する技術です。モデルの再学習なしで最新・専門情報に基づいた正確な回答が可能になり、ハルシネーション（嘘）の低減や情報源の明示に強みがあります。
Source updated 2026-04-28 · Snapshot 2026-10-08
Source links and calculated neighbors
Cosine values measure shared lexical features, not truth, agreement or identical meaning. Original reference links are labeled separately.
- Vocal generationComputed lexical cosine similarity 0.113 · shared title, text, tags and references
- HallucinationComputed lexical cosine similarity 0.102 · shared title, text, tags and references
- anthropicComputed lexical cosine similarity 0.085 · shared title, text, tags and references