A Normative Intermediate Representation for ASP-Based Compliance Reasoning

2026-06-03

AI Summary

This work addresses the challenge of uniformly expressing modal, temporal, and external function constructs within Answer Set Programming (ASP) for compliance reasoning. To this end, the paper proposes MONIR, a modalized output specification intermediate representation tailored for ASP. MONIR introduces a phased operational semantics that supports executable compilation, external function calls, and temporal rules, and it is the first to integrate modal semantics into a specification intermediate representation framework. By leveraging large language models for automated extraction and employing modular incremental solving techniques, the approach enables structured modeling and efficient reasoning over complex regulations, such as China’s ADAS standards. Experimental results demonstrate significant improvements in both specification extraction accuracy and ASP solving efficiency, validating the feasibility and performance gains of modular, incremental reasoning in compliance checking.

Abstract

We propose MONIR, a Modalized-Output Normative Intermediate Representation for ASP-based compliance reasoning. Its core fragment has a staged operational semantics, while MONIR-ASP provides an executable compilation and extensions for external functions, temporal rules, and stable-model reasoning. We instantiate the framework on Chinese ADAS regulations and standards with an LLM-assisted pipeline. Experiments evaluate extraction quality and the efficiency of modular and incremental ASP solving.

Problem

• compliance reasoning
• normative representation
• Answer Set Programming
• intermediate representation
• regulatory compliance

Innovation

• Normative Intermediate Representation
• Answer Set Programming
• Compliance Reasoning
• Modalized Semantics
• LLM-assisted Extraction