Intent-Based Network Management with Large Language Models
Dr. Carlos Bernardos, Prof. Antonio de la Oliva
Universidad Carlos III de Madrid
Abstract
We propose an intent-based network management system powered by large language models that translates operator intents expressed in natural language into network configurations. The system uses a chain of specialized LLM agents for intent parsing, conflict resolution, and configuration generation. Evaluation on 500 real operator intents shows 89% end-to-end accuracy in producing valid configurations, with the system handling complex multi-domain intents spanning RAN, transport, and core networks.
AI Summary
- LLM-powered system translating natural language intents to network configurations.
- 89% accuracy on 500 real operator intents across RAN, transport, and core.
- Chain of specialized LLM agents for parsing, conflict resolution, and generation.
- Handles complex multi-domain intents spanning multiple network layers.
Key Findings
- 1Intent decomposition into sub-intents significantly improves accuracy.
- 2Conflict detection between intents prevents configuration inconsistencies.
- 3The system learns from operator feedback to improve over time.
Industry Implications
Makes network management accessible to non-expert operators.
Reduces configuration errors that cause service outages.
Foundation for autonomous 6G network operations.
Read the Original Paper
Access the full paper on arXiv for complete methodology, results, and references.
Open on arXivRelated Papers
AI Agents for Autonomous Network Operations: Architecture and Evaluation
Rakuten Mobile / NICT Japan — 18 citations
AI/ML PapersLarge Language Models for Automated Network Configuration and Troubleshooting
Bell Labs / Nokia — 24 citations
AI + Network PapersAI-Native Air Interface Design: End-to-End Learning for 6G Physical Layer
University of Stuttgart — 41 citations