EUVIMEDEUVIMED entdecken
Crossref

On the Convergence of Large Language Model Optimizer for Black-Box Network Management

Hoon Lee, Wentao Zhou, Mérouane Debbah, Inkyu Lee

IEEE Transactions on Communications · 2025-11 · 73 (11) · 11385-11402

DOI: 10.1109/tcomm.2025.3592598 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Communications, 2025-11. Verfasst von Hoon Lee und 3 weiteren Autor:innen. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 59 Literaturangaben. Gefördert von: National Research Foundation of Korea, Ministry of Science and ICT (MSIT), Korea Government, Basic Research Program-Individual Research(Global Matching Grants-Sweden) from the NRF.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Future wireless networks are expected to incorporate diverse services that often lack general mathematical models. To address such black-box network management tasks, the large language model (LLM) optimizer framework, which leverages pretrained LLMs as optimization agents, has recently been promoted as a promising solution. This framework utilizes natural language prompts describing the given optimization problems along with past solutions generated by LLMs themselves. As a result, LLMs can obtain efficient solutions autonomously without knowing the mathematical models of the objective functions. Although the viability of the LLM optimizer (LLMO) framework has been studied in various black-box scenarios, it has so far been limited to numerical simulations. For the first time, this paper establishes a theoretical foundation for the LLMO framework. With careful investigations of LLM inference steps, we can interpret the LLMO procedure as a finite-state Markov chain, and prove the convergence of the framework. Our results are extended to a more advanced multiple LLM architecture, where the impact of multiple LLMs is rigorously verified in terms of the convergence rate. Comprehensive numerical simulations validate our theoretical results and provide a deeper understanding of the underlying mechanisms of the LLMO framework.

Förderung

  • National Research Foundation of Korea
  • Ministry of Science and ICT (MSIT), Korea Government
  • Basic Research Program-Individual Research(Global Matching Grants-Sweden) from the NRF
  • MSIT in the year 2024
  • Institute of Information and Communications Technology Planning and Evaluation (IITP) Grants by MSIT
  • Intelligent 6G Wireless Access System
  • Artificial Intelligence Graduate School Program

Zitieren

APA

Lee, H., Zhou, W., Debbah, M., & Lee, I. (2025). On the Convergence of Large Language Model Optimizer for Black-Box Network Management. IEEE Transactions on Communications, 73(11), 11385-11402. https://doi.org/10.1109/tcomm.2025.3592598

Vancouver

Lee H, Zhou W, Debbah M, Lee I. On the Convergence of Large Language Model Optimizer for Black-Box Network Management. IEEE Transactions on Communications. 2025;73(11):11385-11402. doi:10.1109/tcomm.2025.3592598

BibTeX
@article{Lee2025,
  author = {Lee, Hoon and Zhou, Wentao and Debbah, Mérouane and Lee, Inkyu},
  title = {On the Convergence of Large Language Model Optimizer for Black-Box Network Management},
  journal = {IEEE Transactions on Communications},
  year = {2025},
  volume = {73},
  number = {11},
  pages = {11385--11402},
  doi = {10.1109/tcomm.2025.3592598}
}
Kostenlos starten →