Skip to main content
Search Context

Debate on KGs and LLMs

Knowledge Graphs in the era of Large Language Models: a panel debate


The rapid development of Large Language Models (LLMs) has sparked debate about the role and future of Knowledge Graphs (KGs). LLMs offer impressive capabilities in natural language understanding and generation. Many argue that they are already and they will further enhance their ability to understand and reason. Today, LLMs face limitations in delivering structured, precise, and explainable knowledge. Knowledge Graphs, by contrast, provide a robust framework for representing and linking data and knowledge in a structured manner, facilitating interoperability and reasoning.

This debate explores the complementarity and potential conflicts between LLMs and KGs, examining their respective roles in driving advancements in artificial intelligence and knowledge representation. To what degrees are KGs and LLMs complementary? Or will LLMs subsume KG capabilities? Or will KGs be essential to provide context and refine the results of LLMs? What is the implication for our research community?

To explore this topic, we are excited to organize a panel debate where we will have two sides: Team KG and Team LLM. Our debaters will be Paul Groth, Irene Celino, Oscar Corcho, Anna Lisa Gentile. Who will take which side? What will their arguments be? You will have to come to the panel debate!

Panelists


Paul Groth
http://pgroth.com/bio/

Paul Groth is Professor of Algorithmic Data Science at the University of Amsterdam where he leads the Intelligent Data Engineering Lab (INDElab). He holds a Ph.D. in Computer Science from the University of Southampton (2007) and has done research at the University of Southern California, the Vrije Universiteit Amsterdam and Elsevier Labs. His research focuses on intelligent systems for dealing with large amounts of diverse contextualized knowledge with a particular focus on web and science applications. This includes research in data provenance, data integration and knowledge sharing. Paul is scientific director of the UvA’s Data Science Center. Additionally, he is co-scientific director of two Innovation Center for Artificial Intelligence (ICAI) labs: The AI for Retail (AIR) Lab – a collaboration between UvA and Ahold Delhaize; and the Discovery Lab – a collaboration between Elsevier, the University of Amsterdam and VU University Amsterdam. Previously, Paul led the design of a number of large scale data integration and knowledge graph construction efforts in the biomedical domain. Paul was co-chair of the W3C Provenance Working Group that created a standard for provenance interchange. He has also contributed to the emergence of community initiatives to build a better scholarly ecosystem including altmetrics and the FAIR data principles. Paul is co-author of “Provenance: an Introduction to PROV” and “The Semantic Web Primer: 3rd Edition” as well as numerous academic articles. You can find him on twitter: @pgroth.


Irene Celino
https://iricelino.org/

Irene Celino is a Computer Scientist with a passion for Research and Innovation. She is an expert in Web and Data technologies, with a specialization in Semantic Web, Semantic Interoperability, Knowledge Graphs, Data Space and Data Fabric, Human acceptance of AI, Human Computation, Citizen Science, Human-in-the-loop Machine Learning and Data Analytics. She is the Head of the Knowledge Technologies group, one of the 8 competence centers in Cefriel, the research, innovation and technology transfer center of Politecnico di Milano, where she leads a group of talented professionals specialized in Knowledge Graphs and Human Computation with a focus on Smart Cities, Mobility/Transport and Industry 5.0. Irene has ~20 years of experience in 40+ cooperative research projects, at regional, national and European level; She started as junior researcher and she has gained increasing responsibilities over time, and now she also covers the roles of Project Manager, Project Coordinator and Principal Investigator. Irene has been involved in numerous competitive project proposals’ submission, with successful grant acquisition for around 8M euro funding for her organization. She is the author of 100+ publications and has had the pleasure of winning different awards for her scientific and applied research work. Among other research roles, Irene is a member of the Editorial Board of the Open Access Transactions on Graph Data and Knowledge (TGDK).


Annalisa Gentile
https://w3id.org/people/annalisa

Annalisa is a senior research scientist at IBM Research Almaden, USA. Her main Research Areas are Information Extraction (IE), Natural Language Processing (NLP) and Semantic Web.

From December 2015 to February 2017 Annalisa was a post-doctoral research scientist at the University of Mannheim. From May 2010 to November 2015 she was a post-doctoral research associate at the University of Sheffield. Annalisa obtained my doctoral degree with a thesis on Named Entity Disambiguation at the University of Bari, Italy in 2010.


Oscar Corcho
https://oeg.fi.upm.es/index.php/en/teachers/11-ocorcho/

Politécnica de Madrid. His core research activities are centered around Open Science, Knowledge-Graph-based Data Integration and Ontological Engineering. These are applied to several domains of expertise, with a strong focus on the Public Sector, where Oscar has worked in a large number of open data and data spaces initiatives. He is one of the academic directors of the AI4Gov (Artificial Intelligence for Public Services) joint Master between UPM and POLIMI, and the director of the UPM chair on data governance funded by the City of Zaragoza, under which he has contributed to the data governance ordnance template by the Spanish Federation of Municipalities and Provinces.