KG-NeSy 2025

Workshop Program

• 15:00–15:05 — Opening (~5 minutes)

• 15:05–15:50 — Keynote
  Cogan Shimizu, Wright State University
  Title: Accelerating Knowledge Engineering with Modularity
  Abstract: Large Language Models show promise to significantly accelerate key Knowledge Graph and Ontology Engineering tasks, including ontology modeling, extension, modification, population, alignment, and entity disambiguation. Yet, techniques to accomplish these tasks are (currently) of high variance in effectiveness. We previously laid out how modular approaches to ontologies will be of central importance. This talk confirms and updates aspects to leveraging modularity (and patterns) for increased effectiveness.

• 15:50–16:10 — Intermediate Languages Matter: Formal Languages and LLMs affect Neurosymbolic Reasoning.
  Alexander Beiser, Nysret Musliu, David Penz

• 16:10–16:30 — Semantic-Driven Data Augmentation for Improved Machine Learning Predictions (Extended Abstract).
  Majlinda Llugiqi, Fajar J. Ekaputra, Marta Sabou

• 16:30–16:40 — Break (~10 minutes)

• 16:40–17:00 — Evaluating Large Language Models on OWL Lite Reasoning.
  Emanuele Damiano, Francesco Orciuoli

• 17:00–17:20 — Pattern-based AI Risk Assessment: A Taxonomy Expansion Use Case.
  Muhammad Ikhsan, Elmar Kiesling, Salma Mahmoud, Alexander Prock, Artem Revenko, Fajar J. Ekaputra

• 17:20–17:40 — Explainable Zero-Shot Visual Question Answering Logic-Based Reasoning (Extended Abstract).
  Thomas Eiter, Jan Hadl, Nelson Higuera, Lukas Lange, Johannes, Bileam Scheuvens, Jannik Strötgen

• 17:40–18:00 — Guiding LLM Generated Mappings with Lifecycle-Based Metadata: An Early Evaluation.
  Sarah Alzahrani, Declan O'Sullivan

• 18:00 — Closing

Call for Papers

Knowledge Graphs provide a structured and robust framework for organizing, integrating, and reasoning over large-scale knowledge in machine-readable formats. Due to their ability to seamlessly integrate structured symbolic knowledge with data-driven techniques, they are increasingly used in combination with Machine Learning methods. This interplay has led to significant advancements in AI applications, paving the way for the emerging paradigm of Neurosymbolic AI, which seeks to unify neural learning models with symbolic reasoning.

We welcome submissions that explore these synergies as well as any other combinations, seeking to gain a better understanding of how Knowledge Graphs and Neurosymbolic AI influence and benefit each other. Furthermore, we invite contributions presenting preliminary ideas, in-progress research, and application-focused work. The workshop offers the opportunity to present the work and should spark dialogue and collaboration among participants. 

Topics of Interest

• Neurosymbolic AI for Knowledge Engineering

  • Knowledge Representation and Reasoning using Deep Neural Networks

  • Large-Language Models for Knowledge Engineering

  • Machine Learning techniques for creating, improving, or aligning Knowledge Graphs.

•   Knowledge Graphs and Neurosymbolic approaches for Robust, Trustworthy and Interpretable AI

  • Knowledge Graph-based approaches for improving fairness, bias mitigation, and ethical AI

  • Knowledge Graphs for trustworthy Neurosymbolic AI systems

  • Knowledge Graphs quality and its influence on Neurosymbolic AI systems

  • Knowledge Infusion in Machine Learning algorithms

  • Neurosymbolic AI for (Autonomous) Agentic Systems

  • Symbolic AI methods, systems, and techniques for Explainable AI

  • Using symbolic reasoning and ontologies to identify, mitigate, and explain biases in neural systems

  • Utilizing knowledge graphs, ontologies, and other structured knowledge representations to enhance AI reliability and accountability

• Application of Neurosymbolic AI and Knowledge Graphs

  • Applications of Neurosymbolic AI and Knowledge Graphs in Industry

  • Applications of Neurosymbolic AI in domains such as medicine, biology, IoT, security, robotics and others

  • Certification, auditing, and documentation of AI systems using Knowledge Graphs and Neurosymbolic AI

  • Integration of Knowledge Graphs, Neurosymbolic AI and human-machine intelligence

Important Dates

• Paper submission deadline: July 4, 2025 (11:59 pm, AoE) - EXTENDED

• Notification of acceptance: August 1, 2025 (11:59 pm, AoE)

• Workshop: September 3, 2025

Organizers

PC Members

• Fariz Darari (University of Indonesia, Indonesia)

• Tobias Geibinger (TU Wien, Austria)

• Filip Ilievski (Vrije Universiteit Amsterdam, The Netherlands)

• Jan-Cristoph Kalo (University of Amsterdam, The Netherlands)

• Fabian Kovac (St. Pölten University of Applied Sciences, Austria)

• Kabul Kurniawan (Gadjah Mada University, Indonesia)

• Majlinda Llugiqi (WU, Austria)

• Diego Rincon-Yanez (Trinity College Dublin, ADAPT Center, Ireland)

• Marta Sabou (WU, Austria)

• Simon Tjoa (St. Pölten University of Applied Sciences, Austria)

Previous Editions

• The 1st edition of KG-NeSy (2024), Innsbruck - Austria (website)

Supported by