Bridging Language Models and Formal Methods for Intent-Driven Optical Network Design

2025-09-26

AI Summary

Large language models (LLMs) inherently suffer from ambiguity and lack of formal semantics when parsing natural-language intents in Intent-Based Networking (IBN), especially in optical networks. Method: This paper proposes the first hybrid framework integrating LLMs, optics-domain retrieval-augmented generation (RAG), and formal verification. It leverages LLMs for initial informal intent parsing, employs optics-specific RAG to inject domain knowledge and constrain semantic interpretation, and applies symbolic reasoning with formal verification—grounded in optical networking standards (e.g., ITU-T G.698.x)—to ensure syntactic correctness and protocol compliance of generated topologies. Contribution/Results: Compared to LLM-only approaches, our framework significantly improves accuracy and verifiability of intent-to-optical-topology mapping. It produces end-to-end interpretable, formally verifiable, and standards-compliant configuration plans, establishing a novel paradigm for highly reliable IBN design in optical infrastructure.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

Intent-Based Networking (IBN) aims to simplify network management by enabling users to specify high-level goals that drive automated network design and configuration. However, translating informal natural-language intents into formally correct optical network topologies remains challenging due to inherent ambiguity and lack of rigor in Large Language Models (LLMs). To address this, we propose a novel hybrid pipeline that integrates LLM-based intent parsing, formal methods, and Optical Retrieval-Augmented Generation (RAG). By enriching design decisions with domain-specific optical standards and systematically incorporating symbolic reasoning and verification techniques, our pipeline generates explainable, verifiable, and trustworthy optical network designs. This approach significantly advances IBN by ensuring reliability and correctness, essential for mission-critical networking tasks.

Problem

• Translating informal natural-language intents into formal optical topologies
• Addressing ambiguity and lack of rigor in Large Language Models
• Generating explainable and verifiable optical network designs

Innovation

• Integrates LLM-based intent parsing with formal methods
• Uses Optical RAG for domain-specific optical standards
• Applies symbolic reasoning for verifiable network designs