Tuesday May 21, 2024

09:00 - 10:30    Session 1 Chair: Hugo Gonçalo Oliveira, University of Coimbra, Portugal

9:00 - 9:20    Investigating the Impact of Different Graph Representations for Relation Extraction with Graph Neural Networks Moritz Blum1, Gennaro Nolano2, Basil Ell3, Philipp Cimiano4 1Bielefeld University, 2Mare Group, 3Bielefeld University, University of Oslo, 4Univ. Bielefeld

9:20 - 9:40    TaxoCritic: Exploring Credit Assignment in Taxonomy Induction with Multi-Critic Reinforcement Learning Injy Sarhan1, Bendegúz Toth2, Pablo Mosteiro2, Shihan Wang2 1Shell / Utrecht University, 2Utrecht University

9:40 - 10:00    Combining Deep Learning Models and Lexical Linked Data: Some Insights from the Development of a Multilingual News Named Entity Recognition and Linking Dataset Emmanuel Cartier and Emile Peetermans European Commission, Joint Research Center

10:00 - 10:20    Deductive Verification of LLM Generated SPARQL Queries Alexandre Rademaker1, Guilherme Lima2, Sandro Rama Fiorini3, Viviane Torres da Silva3 1IBM Research and EMAp/FGV, 2IBM Research Brazil, 3IBM Research

10:30    Coffee break

11:00 - 13:00    Session 2 Chair: Gilles Sérasset, Université Grenoble Alpes, France

11:00 - 11:20    How to Turn Card Catalogs into LLM Fodder Mary Ann Tan, Shufan Jiang, Harald Sack FIZ Karlsruhe

11:20 - 11:40    Evaluating Large Language Models for Linguistic Linked Data Generation Maria Pia di Buono1, Blerina Spahiu2, Verginica Barbu Mititelu3 1University of Naples "L'Orientale", 2University of Milan - Bicocca, 3RACAI

11:40 - 12:00    Towards Automated Evaluation of Knowledge Encoded in Large Language Models Bruno Carlos Luís Ferreira, Catarina Silva, Hugo Gonçalo Oliveira University of Coimbra

12:00 - 12:20    Self-Evaluation of Generative AI Prompts for Linguistic Linked Open Data Modelling in Diachronic Analysis Florentina Armaselu1, Chaya Liebeskind2, Giedre Valunaite Oleskeviciene3 1University of Luxembourg, 2Jerusalem College of Technology, 3Mykolas Romeris University

12:20 - 13:00    Discussions and further work on the topic