Registration

Please register the workshop at registration page of JSAI International Symposia on AI 2026.

Topics

Relevant topics include, but are not limited to, the following:

• Model of legal reasoning
• Argumentation / Negotiation / Argumentation agent
• Legal term ontology
• Formal legal knowledge-base / Intelligent management of legal knowledge-base
• Computer-aided legal education
• Use of informatics and AI in law
• Social implications of use of informatics and AI in law
• AI and intellectual property
• Legal/Ethical Compliance check of AI Systems
• Natural language processing for legal knowledge
• Translating law into formal representation
• Legal data mining
• Legal document analysis
• Verification and validation of legal knowledge systems
• Online dispute resolution
• Evidential reasoning
• AI application to forensics
• AI application to smart contracts and blockchain
• Legislation support by AI/IT techniques
• Generative AI and Law
• Any theories and technologies which is not directly related with juris-informatics but has a potential to contribute to this domain

Invited Speakers

Ryohei Hisano, University of Tokyo, Japan
Title: Structuring and Comparing Legal Knowledge through Knowledge Graphs and Dynamic Networks
Abstract:
Legal knowledge is not a static collection of statutes, precedents, and concepts. It is a dynamic and structured body of reasoning that connects facts, norms, legal interpretations, institutional practices, and jurisdiction specific contexts. Building reliable legal AI therefore requires more than semantic search or text generation. It requires methods for making legal knowledge explicit, formalizing tacit professional knowledge, and analyzing how legal norms and legal concepts evolve over time. In this talk, I will discuss our recent work on legal knowledge structuring for interpretable legal AI and legal data science. The first part focuses on ontology based and knowledge graph based representations of legal reasoning. By representing relations among facts, legal norms, statutes, precedents, interpretations, and conclusions, these structures transform implicit legal and institutional reasoning into machine readable, AI ready knowledge. This provides a foundation for transparent legal information retrieval, explanation of legal reasoning, human in the loop knowledge construction, and collaboration between legal experts and AI systems. The second part turns to the dynamics of legal knowledge. Legal systems evolve through statutory interpretation, judicial reasoning, doctrinal refinement, and institutional practice. To capture these processes, we develop interpretable methods for dynamic networks and hypergraphs, including change point detection and geometric analysis of network trajectories. These methods represent legal knowledge networks as trajectories, decompose temporal variation into interpretable modes, and attribute observed changes to individual nodes, providing quantitative evidence for legal experts to examine structural change over time. Finally, I will connect these ideas to our JSPS–SNSF Japan–Switzerland Joint Research Project with Prof. Dr. Tilmann Altwicker at the University of Zurich. The project develops interoperable legal ontologies and graph based methods for comparing legal knowledge across jurisdictions, including Japan, Switzerland, Germany, the United States, and international courts. It also examines how courts contribute to legal development and how differences between common law and civil law systems change over time.
Bio:
Ryohei Hisano is a Lecturer at the University of Tokyo and a Principal Researcher at the Canon Institute for Global Studies. His research lies at the intersection of mathematical informatics, network science, semantic web technologies, and legal data science, with broader applications in finance, public administration, and institutional decision making. His work develops graph based and ontology based methods for structuring complex legal and social knowledge. In the legal domain, he studies how legal norms, facts, judicial reasoning, precedents, statutes, and institutional practices can be represented as interoperable knowledge structures. His recent research includes legal knowledge graph construction, ontology based modeling of legal reasoning, and dynamic network methods for analyzing the evolution and consistency of legal norms across time and jurisdictions. He received his PhD from ETH Zurich and has held research positions at ETH Zurich, the National Institute of Informatics, and the University of Tokyo. He is the Japanese side Principal Investigator of the JSPS–SNSF Japan–Switzerland Joint Research Project, “Global Framework for Legal Knowledge Structuring and AI Development,” together with Prof. Dr. Tilmann Altwicker at the University of Zurich. The project extends this line of work through comparative legal ontologies, graph based analysis of norm evolution, and quantitative methods for studying legal development across jurisdictions.

Le-Minh Nguyen, Japan Advanced Institute of Science and Technology, Japan
Title: Legal Case Retrieval and Entailment in the Era of Agentic LLMs
Abstract:
Legal case retrieval and legal entailment are fundamental tasks for building intelligent legal information systems. They require not only matching queries with relevant precedents, but also understanding whether a cited case, statute, or legal proposition supports, contradicts, or entails a given legal conclusion. Recent advances in Large Language Models have significantly improved semantic understanding, reasoning, and natural language generation; however, legal reasoning remains challenging because it demands factual accuracy, citation faithfulness, interpretability, and careful handling of jurisdiction-specific knowledge. In this talk, I will discuss recent progress in legal case retrieval and entailment, with a focus on how Agentic LLMs can transform legal text processing. Unlike conventional LLM-based systems that generate answers in a single step, agentic systems can plan, retrieve, verify, compare, and revise their reasoning through interactions with external tools, legal databases, and structured knowledge sources. This opens new possibilities for multi-step legal research, precedent analysis, and explainable legal decision support. I will also highlight key challenges, including hallucination, evaluation of legal reasoning, robustness across legal domains, and the need for trustworthy retrieval-augmented generation. Finally, the talk will outline future directions toward reliable, interpretable, and human-centered legal AI systems that combine neural language models, symbolic reasoning, and agentic workflows.
Bio:
Le-Minh Nguyen is a Professor in the School of Information Science at JAIST and the former Director of the Interpretable AI Center. He leads the Machine Learning and Natural Language Understanding Laboratory at JAIST and recently completed a sabbatical at Imperial College London, UK (May 2025–April 2026). His research interests include machine learning, deep learning, natural language processing, legal text processing, and explainable AI (XAI). He serves as an action editor of TACL (a leading journal in NLP), a board member of VLSP (Vietnamese language and speech processing), and an editorial board member of AI & Law and the Journal of Natural Language Processing (Cambridge). He is also a steering committee member of Juris-informatics (Jurisin) in Japan—a research area that examines legal issues from an informatics perspective. He received the JSAI Best Paper Award in 2025, the SAC Highlight Award at ACL 2025, and his work on efficient LLMs was also nominated for Best Paper at ACL 2026.

Submissions

We welcome and encourage the submission of high quality, original papers, which are not simultaneously submitted for publication elsewhere. Papers should be written in English, formatted according to the Springer Verlag LNCS style in a pdf form, which can be obtained from https://www.springer.com/gp/computer-science/lncs/conference-proceedings-guidelines and not exceed 14 pages including figures, references, etc. If you use a word file, please follow the instruction of the format, and then convert it into a pdf form and submit it at the paper submission page:
https://easychair.org/conferences/?conf=jurisin2026.

If you cannot submit a paper by EasyChair System by some trouble, please send email to "ksatoh[at]nii.ac.jp"

If a paper is accepted, at least one author of the paper must register the workshop through this page. Without fulfilling this condition, the paper will not be in the proceedings.

Back To Top