🗓 Program
| 09:00 – 10:00 | Opening and Invited Talk |
| 10:00 – 10:40 | ☕ Coffee Break |
Session 1: Structured Knowledge Meets Language Models
Chairs: Matteo Magnini and Elena Umili — 10:40 – 12:40
| Time | Paper | Authors |
|---|---|---|
| 10:40 – 11:00 | A Semantic Knowledge Graph Construction Pipeline for Vehicle Intention Prediction in nuScenes | Khaled Nowara (German University in Cairo), Mohamed Manzour (Computer Engineering Department, University of Alcalá, Madrid, Spain), Catherine Elias (German University in Cairo) |
| 11:00 – 11:20 | Automatic Chain of Concepts: Conceptual Prompting for LLMs by Constructing Concept Trees | Hurile Borjigin (Technical University of Munich, Siemens AG, Munich, Germany), Nishtha N. Vaidya (Technical University of Munich, Siemens AG, Munich, Germany), Thomas A. Runkler (Technical University of Munich, Siemens AG, Munich, Germany), Veronika Haderlein-Høgberg (Siemens AG, Munich, Germany) |
| 11:20 – 11:40 | Identifying Semantic Gaps between Legal Case Descriptions and Logical Fact Formulas Using LLMs: Definition and Preliminary Evaluation | Wachara Funwacharakorn (Center for Juris-Informatics, ROIS-DS, Tokyo, Japan), May Myo Zin (Center for Juris-Informatics, ROIS-DS, Tokyo, Japan), Ken Satoh (Center for Juris-Informatics, ROIS-DS, Tokyo, Japan) |
| 11:40 – 12:00 | When AI Should Refuse to Decide: Formal Irresolution and the Ethics of Legal Indeterminacy in LLM-Based Reasoning for Biodiversity Law | Rohan Nanda (Law & Tech Lab, Department of Advanced Computing Sciences, Maastricht University), Henrique Marcos (Faculty of Law, Maastricht University), Júlia Schütz Veiga (Ocean Voices Programme, University of Edinburgh) |
| 12:00 – 12:20 | On the Reliability of LLM Orchestration Under Horizontal Knowledge Partitioning | Mayar Abdelwahab (German University in Cairo), Nourhan Ehab (German University in Cairo), Mervat Abuelkheir |
| 12:20 – 12:40 | KGP-QG: Multi-hop Reasoning with Knowledge Graphs in LLMs | Al Hasib Mahamud (University of Lethbridge), Yllias Chali (University of Lethbridge) |
| 12:40 – 14:00 | 🍽 Lunch Break |
Session 2: Neuro-Symbolic Integration: Logic, Belief, and Agent Design
Chair: Mervat Abuelkheir — 14:00 – 16:00
| Time | Paper | Authors |
|---|---|---|
| 14:00 – 14:20 | Neurosymbolic Clinical Reasoning: From Evidence Extraction to Reason-Based Decision | Henrique Marcos (Faculty of Law, Maastricht University), Foivos Ioannis Tzavellos (Graduate school of Informatics, University of Amsterdam), Rohan Nanda (Law and Tech Lab, Department of Advanced Computing Sciences, Maastricht University) |
| 14:20 – 14:40 | Toward a Requirements-Driven Neuro-Symbolic Approach to Developing Trustworthy AI Agents Systems | Bita Banihashemi (IGDORE), Yves Lespérance (York University), Sotirios Liaskos (York University) |
| 14:40 – 15:00 | Epistemically-Constrained Belief Harmonization: Integrating Symbolic Structure with Probabilistic Consensus | Adam Kostka (Warsaw University of Technology), Jarosław A. Chudziak (Warsaw University of Technology) |
| 15:00 – 15:20 | A Parametric Model of Cognitive Complexity for Symbolic Knowledge-Based Decision Systems | Federico Sabbatini (University of Bologna) |
| 15:20 – 15:40 | PrologMCP: A Standardized Prolog Tool Interface for LLM Agents | Agnieszka Mensfelt (Royal Holloway, University of London), Adarsh Prabhakaran (Royal Holloway, University of London), Adrian Haret (Royal Holloway, University of London), Vince Trencsenyi (Royal Holloway University of London, United Kingdom), Kostas Stathis (Royal Holloway, University of London) |
| 15:40 – 16:00 | NL2UNIFOL: from Natural Language Sentences to Uniform First-Order Logic Formulae | Matteo Magnini (University of Luxembourg), Davide Liga (University of Luxembourg), Luca Pasetto (University of Luxembourg) |
| 16:00 – 16:30 | ☕ Coffee Break |
Session 3: Symbolic Constraints in Learning and Inference
Chair: Federico Sabbatini — 16:30 – 17:50
| Time | Paper | Authors |
|---|---|---|
| 16:30 – 16:50 | First Steps Towards Human-AI Ranking Aggregation | Jonas Karge (TU Dresden), Roy Ferguson (University of Cape Town and CAIR), Daniel Grimaldi (University of Cape Town and CAIR), Jonas Haldimann (University of Cape Town and CAIR), Ruvarashe Madzime (University of the Western Cape and CAIR), Thomas Meyer (University of Cape Town and CAIR) |
| 16:50 – 17:10 | Neuro-Symbolic Injection of LTLf Constraints in Autoregressive Reinforcement Learning Policies | Ashkan Ansarifard (Sapienza University of Rome), Matteo Mancanelli (Sapienza University of Rome), Elena Umili (Sapienza University of Rome), Fabio Patrizi (Sapienza University of Rome) |
| 17:10 – 17:30 | Enhancing Retrieval-Augmented Generators with Symbolic Query Definitions | Filippo Bianchini (Sapienza University of Rome), Marco Calamo (Sapienza University of Rome), Marco Console (Sapienza University of Rome), Massimo Mecella (Sapienza University of Rome), Laura Papi (Sapienza University of Rome) |
| 17:30 – 17:50 | Deep Weighted Finite Automata | Francesco Casone (University of Milano-Bicocca), Rafael Peñaloza (University of Milano-Bicocca) |
| 17:50 – 18:00 | Closing |
🎤 Invited Speakers
Giovanni Ciatto
Tenure-track Researcher, University of Bologna
Giovanni Ciatto is a tenure-track researcher at the Department of Computer Science and Engineering, University of Bologna, with a PhD in Data Science and Computation and a Master Degree in Computer Science and Engineering. His research focuses on AI, logic programming, multi-agent systems, distributed systems, and software engineering, with a particular emphasis on symbolic reasoning, sub-symbolic processing, and their intersection. He has taught these topics at the University of Bologna and international venues and has co-authored over 70 publications. He is actively involved in organising special issues, reviewing for major AI and MAS journals and conferences, and contributing to research-oriented open-source software tools, including logic solvers (e.g., 2P-Kt) and BDI agent programming toolkits (e.g., JaKtA).