Automated detection of contradictions in 5G network specifications using reinforcement learning-trained small LLM

Contradictions in telecommunications specifications can lead to implementation errors, interoperability issues, and security vulnerabilities. This paper presents a framework for automating the detection of contradictions in telecommunications standards using a 3B parameter large language model (LLM)...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	EURASIP journal on wireless communications and networking Ročník 2025; číslo 1; s. 85 - 27
Hlavní autoři:	Zhang, Wenbin, Wei, Qiang, Chen, Haowen, Wang, Yunfeng
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Cham Springer International Publishing 27.10.2025 Springer Nature B.V SpringerOpen
Témata:	5G mobile communication 5G specification Accuracy Automation Cognition & reasoning Communications Engineering Communications networks Contradiction detection Cybersecurity Documents Emerging Technologies in Mobile Computing for IoT Connectivity and Edge Computing Engineering Hallucinations Information Systems Applications (incl.Internet) Internet of Things Interoperability Knowledge representation Language Large language models LLM Machine learning Natural language Network security Networks Power efficiency Privacy Reinforcement learning Semantics Signal,Image and Speech Processing Specifications Subject specialists Taxonomy Telecommunications LLM Contradiction detection 5G specification Reinforcement learning
ISSN:	1687-1499, 1687-1472, 1687-1499
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	Contradictions in telecommunications specifications can lead to implementation errors, interoperability issues, and security vulnerabilities. This paper presents a framework for automating the detection of contradictions in telecommunications standards using a 3B parameter large language model (LLM) trained via reinforcement learning. We develop a formal taxonomy of contradiction types in 5G specifications, generate high-quality synthetic training data using carefully designed prompts to larger LLMs, and train a specialized model using group relative policy optimization (GRPO) with a multi-component reward function. Our experiments demonstrate that our RL-trained small model achieves superior performance compared to larger general models on both synthetic and real-world contradictions from the CellularLint dataset. The effectiveness of our approach has significant implications for internet of things (IoT) ecosystems, where consistent and reliable network specifications are crucial for ensuring interoperability, power efficiency, and security across billions of connected devices that rely on 5G infrastructure. Our work demonstrates that specialized smaller models can exceed the capabilities of larger general models on domain-specific tasks, offering a resource-efficient approach to specification verification that could accelerate the development and deployment of reliable IoT systems.
Bibliografie:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ISSN:	1687-1499 1687-1472 1687-1499
DOI:	10.1186/s13638-025-02523-3