Volltext-Downloads (blau) und Frontdoor-Views (grau)

Terminologie und Retrieval- Augmented Generation (TermRAG)

  • In this article, we introduce the concepts of Retrieval-Augmented Generation (RAG) and TermRAG. RAG is a technique that Combines the capabilities of generative language models with information-retrieval methods for more precise and reliable output. TermRAG is a specialized variant of RAG that leverages terminological resources during the retrieval process. We further discuss appiication scenarios of TermRAG in specialized communication and outline the current state of technical implementation options.

This document is embargoed until:

2027/03/01

Export metadata

Additional Services

Search Google Scholar

Statistics

frontdoor_oas
Metadaten
Author:Karolina SuchowolecGND, Christian LangGND
URN:urn:nbn:de:bsz:mh39-136860
ISBN:978-3-7329-1225-4
ISSN:2190-7471
Parent Title (German):Wenn KI auf Fach und Sprache trifft
Series (Serial Number):Studien zu Fach, Sprache und Kultur (12)
Publisher:Frank & Timme
Place of publication:Berlin
Editor:Ines-Andrea Busch-LauerORCiDGND
Document Type:Part of a Book
Language:German
Year of first Publication:2026
Date of Publication (online):2026/03/11
Publishing Institution:Leibniz-Institut für Deutsche Sprache (IDS)
Publicationstate:Zweitveröffentlichung
Reviewstate:(Verlags)-Lektorat
Tag:Retrieval-Augmented Generation (RAG); generative language models; specialized communication; terminological resources
GND Keyword:Generierung <Sprache>; KI
First Page:75
Last Page:90
DDC classes:400 Sprache / 400 Sprache, Linguistik
Open Access?:ja
Program areas:Grammatik
Licence (German):License LogoCreative Commons - CC BY-SA - Namensnennung - Weitergabe unter gleichen Bedingungen 4.0 International