An LLM-Based Framework for Intent-Driven Network Topology Design
2026-07-01 • Networking and Internet Architecture
Networking and Internet ArchitectureArtificial IntelligenceComputation and Language
AI summaryⓘ
The authors studied how well Large Language Models (LLMs) can create network designs that follow specific rules and stay reliable. They built a system combining step-by-step modeling and checks to make sure the generated network maps are correct and meet the constraints. They tested different LLMs on realistic network examples and measured how closely the generated designs matched real ones, and how good they were at keeping servers and content connected. The authors also looked at common mistakes like connection mismatches. Their work helps understand which LLMs are better for automated network design.
Large Language Modelsnetwork topologyconstraint-driven designhierarchical modelingstructural correctnessresilience metricsnode and edge F1-scorenetwork automationconnectivitybenchmarking
Authors
Kholoud El-Habbouli, Fen Zhou, Stephane Huet
Abstract
Designing deployable and resilient network topologies from natural language requirements remains a challenging problem in network automation. This work investigates the ability of Large Language Models (LLMs) to generate structurally valid and constraint-compliant network topologies through a constraint-driven pipeline combining hierarchical modeling and systematic validation. The framework is evaluated via a multimodel comparison of proprietary and open-weight LLMs across four realistic network scenarios released as a public dataset. We assess structural correctness using node and edge F1-scores against reference topologies, and evaluate resilience through server and content connectivity metrics. In addition, we analyze common failure modes, including interface mismatches and directional inconsistencies in generated topologies. Overall, this work provides a systematic benchmark for understanding how LLMs handle structural and resilience constraints in topology synthesis, and supports informed model selection for AI-driven network design.