Horae: A Domain-Agnostic Modeling Language for Automating Multimodal Service Regulation⋆

2024-06-06

2024 IEEE International Conference on Web Services (ICWS)

AI Summary

To address the fragmentation, high manual effort, and cross-domain adaptation challenges in modeling multi-domain, multimodal regulatory rules, this paper proposes Horae—a domain-agnostic, unified multimodal regulatory rule modeling language—and an end-to-end intelligent regulatory framework. The framework integrates a fine-tuned large language model (RuleGPT) with multimodal rule parsing and generation techniques, achieving, for the first time, deep LLM embedding across the entire automated regulatory rule modeling pipeline. Horae enables formal specification and rapid cross-domain transfer of regulatory rules across finance, healthcare, transportation, and other domains. Experimental results demonstrate a substantial improvement in rule modeling efficiency and a reduction of over 70% in cross-domain adaptation time, validating the feasibility and generalizability of end-to-end automated regulatory compliance.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Artificial intelligence is rapidly encroaching on the field of service regulation. This work-in-progress article presents the design principles behind Horae, a unified specification language to model multimodal regulation rules across a diverse set of domains. We show how Horae facilitates an intelligent service regulation pipeline by further exploiting a fine-tuned large language model named RuleGPT that automates the Horae modeling process, thereby yielding an end-to-end framework for fully automated intelligent service regulation.

Problem

• Develops a unified language for multimodal service regulation.
• Automates service regulation using a fine-tuned language model.
• Creates an end-to-end framework for intelligent service automation.

Innovation

• Unified specification language for multimodal regulation
• Fine-tuned large language model automates modeling
• End-to-end framework for intelligent service regulation