associated with 22nd International Conference on Principles of Knowledge Representation and Reasoning (KR 2025)
The aim of this workshop is to bring together researchers and practitioners working on the intersections of language models, knowledge representation, and reasoning, particularly, but not exclusively, in medical, law, and science domains. We encourage submissions that discuss novel techniques, approaches, and innovative ideas related to this topic. Topics of interest include, but are not limited to, the following:
Contributions may be regular papers (up to 9 pages) or short/position papers (up to 5 pages), including abstract, figures, and appendices (if any) but excluding references and acknowledgements. Submissions should follow the KR 2025 formatting guidelines and be submitted through the submission page. Each submission will be reviewed by at least two program committee members.
Single-track schedule. All times are local.
| Time | Item |
|---|---|
| 08:30–09:00 | Registration and Welcome Coffee |
| 09:00–09:15 | Opening Remarks |
| Session 1: Long Papers | |
| 09:15–09:45 | Argumentative Reasoning with Language Models on Non-factorized Case Bases (W. Fungwacharakorn, M. M. Zin, H.-T. Nguyen, Y. Kong, K. Satoh) |
| 09:45–10:15 | Specific Domain Ontology Construction Using Large Language Models (V. M. A. Soares, R. Wassermann) |
| 10:15–10:45 | Which Neurons Nudge Moral Judgments? Neuron-Level Last-Token Steering in Large Language Models (D. Liga, L. Yu, R. Markovich) |
| 10:45–11:00 | Coffee Break |
| Session 2: Long Papers | |
| 11:00–11:30 | Multi-Agent Legal Verifier Systems for Data Transfer Planning (H.-T. Nguyen, W. Fungwacharakorn, K. Satoh) |
| 11:30–12:00 | HA-GNN: Learning Acyclic Hierarchies in Scientific Text with Hierarchy-Aware Graph Neural Networks (D. Joshi, I. Rekik) |
| 12:00–12:30 | Towards a Common Framework for Autoformalization (A. Mensfelt, D. T. Cucala, S. Franco, A. Koutsoukou-Argyraki, V. Trencsenyi, K. Stathis) |
| 12:30–13:30 | Lunch Break |
| Time | Item |
|---|---|
| Invited Talk | |
| 13:30–14:30 | From Classical Agents to LLM Agents: Case Studies from Our Recent Projects (H. Hiyashi) |
| Session 4: Short Papers | |
| 14:30–14:50 | Applying Relation Extraction and Graph Matching to Answering Multiple Choice Questions (N. Shimoda, A. Yamamoto) |
| 14:50–15:10 | FC-CONAN: An Exhaustively Paired Dataset for Robust Evaluation of Retrieval Systems, Exposing the Lower-Bound Bias of Sparse Labels (J. Junqueras, F. Boudin, M. M. Zin, H.-T. Nguyen, W. Fungwacharakorn, D. A. Furman, A. Aizawa, K. Satoh) |
| 15:10–15:30 | On the Role of Domain Experts in LLM-based Knowledge Formalization (S. Vandevelde) |
| 15:30–15:50 | From Data Logs to Narrative Intelligence: Integrating ASP Logic and LLMs for Multi-robot Mission (S. Madasamy, K. Sammut, R. Leibbrandt, P. Santos, C. Buche, V. Martin, A. G. Bosser) |
| 15:50–16:10 | Coffee Break |
| Session 5: Discussion and Closing | |
| 16:10–16:50 | General Discussion and Q&A with Authors |
| 16:50–17:00 | Closing Remarks |
Reasoning is an essential component of human intelligence as it plays a fundamental role in our ability to think critically, support responsible decisions, and solve challenging problems. Traditionally, AI has addressed reasoning in the context of logic-based representations of knowledge. However, the recent leap forward in natural language processing, with the emergence of language models based on transformers, is hinting at the possibility that these models exhibit reasoning abilities, particularly as they grow in size and are trained on more data. Despite ongoing discussions about what reasoning is in language models, it is still not easy to pin down to what extent these models are actually capable of reasoning.
The goal of this workshop is to create a platform for researchers from different disciplines and/or AI perspectives, to explore approaches and techniques with the aim to reconcile reasoning between language models using transformers and using logic-based representations. The specific objectives include analyzing the reasoning abilities of language models measured alongside KR methods, injecting KR-style reasoning abilities into language models (including by neuro-symbolic means), and formalizing the kind of reasoning language models carry out. This exploration aims to uncover how language models can effectively integrate and leverage knowledge and reasoning with it, thus improving their application and utility in areas where precision and reliability are a key requirement.
Prof. Hisashi Hayashi, Ph.D.
Advanced Institute of Industrial Technology (AIIT), Tokyo, Japan
In this invited talk, Prof. Hayashi presents the evolution of agent research from classical autonomous and multi-agent systems to emerging forms of “agentic AI” driven by large language models (LLMs). The first part reviews recent work integrating an online hierarchical task network (HTN) planner with legal and ethical compliance modules, ensuring real-time adherence to frameworks such as the EU AI Act and GDPR in international data transfer and AI utilization contexts.
The second part introduces two LLM-based projects: (1) a framework for AI governance risk scenario generation and mitigation, combining human expertise and generative AI to uncover and structure “unknown unknowns” for proactive risk management; and (2) a multi-agent conversational simulation for disaster evacuation, using LLMs to model collective decision-making and behavioral optimization via natural language analysis.
These studies highlight the transformation of agent research—from symbolic reasoning toward language-based coordination—and suggest directions for building explainable, socially embedded Agentic AI systems.
Prof. Hisashi Hayashi is a professor at the Advanced Institute of Industrial Technology (AIIT) in Tokyo, Japan. His primary research interests include agents and artificial intelligence. He holds a B.E. from Waseda University (Tokyo, Japan) and an M.Sc. and Ph.D. from Imperial College London (U.K.). From 1998 to 1999, he served as a research assistant at Imperial College London, and from 2000 to 2017, he conducted research at the Toshiba R&D Center. Since 2017, he has been with AIIT. Prof. Hayashi is a member of several professional societies, including IPSJ, IEICE, JSAI, JSSST, IIAI, KES, INSTICC, AAAI, and ACM.