Menu
Open
Close

Knowledge Graphs and (Generative) Language Models

Our objective is to integrate external taxonomies, ontologies, or knowledge graphs into language models to develop hybrid architectures that exploit the strengths of both technologies, effectively addressing their limitations for complex tasks.
Back to all topics

Videos, Slides and Posters

Related Publications

Taxonomy-Enhanced Document Retrieval with Dense Representations

Type: Conference Paper

Venue: iDSC 2023

Authors: Victor Mireles, Artem Revenko, Ioanna Lytra, Anna Breit, and Julia Klezl

View publication

Proposal for PORQUE, a Polylingual Hybrid Question Answering System

Type: Poster

Venue: SEMANTiCS 2022  

Authors: Victor Mireles, Artem Revenko, Nikit Srivastava, Daniel Vollmers, Anna Breit and Diego Moussallem

View publication

Related Projects

Logo placeholder

FAIR-AI: Fostering Austria’s Innovative strength and Research excellence in Artificial Intelligence

ongoing
Jan 1, 2024 Dec 31, 2026

The FAIR-AI project aims to address gaps in dealing with societal risks in AI applications, focusing on the challenges posed by the upcoming European AI law.

GeoCROW: GeoSemantic and Crowdsourced enhanced Virtual Reality for Situational Awareness

GeoCROW: GeoSemantic and Crowdsourced enhanced Virtual Reality for Situational Awareness

ongoing
Feb 1, 2023 Jan 31, 2025

GeoCrow project aims to help military forces by conducting disaster relief and emergency operations by providing relevant information about future operational areas through semantically enriched, categorised data, and their visualisation in a virtual reality environment.

Visit Project Website
PORQUE: Polylingual Hybrid Question Answering 1

PORQUE: Polylingual Hybrid Question Answering

finished
Dec 1, 2020 Nov 30, 2023

PORQUE will develop and evaluate an extensible platform for polylingual question answering by relying on knowledge graphs and text. We will deliver a prototype, which will be evaluated through commercial solutions in three domains (open data, publishing, energy).

Visit Project Website
obaris logo

OBARIS: Ontology-Based Artificial Intelligence in Environmental Sector

finished
Feb 1, 2020 Oct 31, 2022

The FFG-funded OBARIS project aims to advance the status quo in the area of auditable semantic artificial intelligence systems (SAISs), by investigating both conceptual aspects of these systems as well as a technology stack that facilitates transposing these system types into concrete settings.

Visit Project Website