Large Language Models Meet Bayesian Optimization: Towards Self-evolving Wireless Digital Twins
Abstract
Wireless digital twins (DTs) are increasingly integrated with artificial intelligence to support online modeling, prediction, and network optimization. However, most existing approaches perform adaptation within predefined learning or optimization structures, which may become unsuitable when wireless tasks and operating conditions change. In this article, we investigate the integration of large language models (LLMs) and Bayesian optimization (BO) to extend wireless DTs from online adaptation towards algorithm-level self-evolution. We introduce an LLM-assisted BO framework in which BO provides a stable and sample-efficient optimization backbone, while the LLM determines whether selected BO components should be retained or adapted based on the current task and accumulated optimization experience. In simulations, we apply the framework to Gaussian process (GP) kernel discovery for a multi-user wireless beam-selection problem. Simulation results show that the LLM-discovered kernel improves the final BO performance by 19.7% on the training tasks compared with the strongest standard kernel baseline. The results demonstrate the potential of combining LLM flexibility with structured BO for self-evolving wireless DTs.
Related
- Bayesian-Guided Cooperative RL Beamforming for Wireless Adversarial User Detection
- Agentic Wireless Digital Twin Construction and Calibration Using Real-World Measurements
- Digital Twin-Guided Wireless Localization via Probabilistic Inverse Learning
- Bayesian Consensus Calibration of Continuously Evolving IRT Item Banks
- Tri-Band Channel Measurement-Enabled Multi-Layer Digital Twin for Terahertz Wireless Data Centers
BibTeX
@misc{sun2026large,
title = {{Large Language Models Meet Bayesian Optimization: Towards Self-evolving Wireless Digital Twins}},
author = {Yingzhuo Sun and Haoyun Li and Ming Xiao and Francisco J. Ferrao and Berna Ozbek},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.31192},
eprint = {2610.31192},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.31192},
}
RIS
TY - UNPB
TI - Large Language Models Meet Bayesian Optimization: Towards Self-evolving Wireless Digital Twins
AU - Yingzhuo Sun
AU - Haoyun Li
AU - Ming Xiao
AU - Francisco J. Ferrao
AU - Berna Ozbek
PY - 2026
DA - 2026/10/22
PB - arXiv
DO - 10.48550/arxiv.2610.31192
UR - https://arxiv.org/abs/2610.31192
AB - Wireless digital twins (DTs) are increasingly integrated with artificial intelligence to support online modeling, prediction, and network optimization. However, most existing approaches perform adaptation within predefined learning or optimization structures, which may become unsuitable when wireless tasks and operating conditions change. In this article, we investigate the integration of large language models (LLMs) and Bayesian optimization (BO) to extend wireless DTs from online adaptation towards algorithm-level self-evolution. We introduce an LLM-assisted BO framework in which BO provides a stable and sample-efficient optimization backbone, while the LLM determines whether selected BO components should be retained or adapted based on the current task and accumulated optimization experience. In simulations, we apply the framework to Gaussian process (GP) kernel discovery for a multi-user wireless beam-selection problem. Simulation results show that the LLM-discovered kernel improves the final BO performance by 19.7% on the training tasks compared with the strongest standard kernel baseline. The results demonstrate the potential of combining LLM flexibility with structured BO for self-evolving wireless DTs.
ER -