NL interface of PROLEG system and extracting normative sentences from German traffic judgements

ABSTRACT

I will give two topics related to NLP and logical inference. (1) NL interface of PROLEG system. We have developed a PROLOG-based legal support system (PROLEG) which simulates judge's reasoning. However, the input from a user are logical fact formulas which the user would find difficult to construct from the current case. We use LLM technology for a front-end of PROLEG which translates NL description into logical fact formulas. (2) Leveraging LLM for Identification and Extraction of Normative Statements. The development of autonomous vehicles (AVs) requires a comprehensive understanding of both explicit and implicit traffic rules to ensure legal compliance and safety. While explicit traffic laws are well-defined in statutes and regulations, implicit rules derived from judicial interpretations and case law are more nuanced and challenging to extract. This research investigates the potential of Large Language Models (LLMs), particularly GPT-4o, in automating the extraction of implicit traffic rules from judicial decisions. By utilizing various prompt engineering techniques, including Standard Prompts, Chain-of-Thought (CoT), Chain-of- Instructions (CoI), and Layer-of-Thought (LoT) prompts, this study aims to assess the effectiveness of GPT-4o in identifying normative content relevant to specific traffic laws. The contributions of this paper include an assessment of LLMs for legal text processing, the automation of implicit rule extraction, and the development of a scalable framework that can continuously update as new  legal precedents emerge. The results indicate promising avenues for integrating automated normative extraction in AV systems, improving both the safety and legal compliance of autonomous driving technologies.

SHORT BIO (from FFJ):

Ken Satoh is a full professor of Principles of Informatics Research Division, NII (National Institute of Informatics), and Sokendai (The Graduate University of Advanced Studies), Japan. He worked with Fujitsu 1981-1995 and then was an associate professor of Hokkaido University till 2001. He also studied law at the law school of University of Tokyo in 2006-2009 and passed Japanese bar exam in 2017. His main research interest is logical foundations of Artificial Intelligence (AI) and application of AI to law. He recently proposed juris-informatics which is a new research field combining informatics and law.

Research themes: AI and Law, Logical Foundation of AI, Compliance Mechanism of AI, Legal Debugging