EUVIMEDEUVIMED entdecken
Crossref

Covert Prompt Transmission for Secure Large Language Model Services

Ruichen Zhang, Yinqiu Liu, Shunpu Tang, Jiacheng Wang, Dusit Niyato, Geng Sun, Yonghui Li, Sumei Sun

IEEE Journal on Selected Areas in Communications · 2026 · 44 · 1589-1603

DOI: 10.1109/jsac.2025.3647412 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Journal on Selected Areas in Communications, 2026. Verfasst von Ruichen Zhang und 7 weiteren Autor:innen. Die Arbeit wurde bisher 30-mal zitiert (Crossref) und nennt 49 Literaturangaben. Gefördert von: Seatrium New Energy Laboratory; in part by Singapore Ministry of Education Tier 1, Nanyang Technological University Centre for Computational Technologies in Finance, Research Innovation and Enterprise 2025 Industry Alignment Fund, Industry Collaboration Projects, administered by the Agency for Science, Technology and Resear….

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper investigates covert prompt transmission for secure and efficient large language model (LLM) services over wireless networks. We formulate a latency minimization problem under fidelity and detectability constraints to ensure confidential and covert communication by jointly optimizing the transmit power and prompt compression ratio. To solve this problem, we first propose a prompt compression and encryption (PCAE) framework, performing surprisal-guided compression followed by lightweight permutation-based encryption. Specifically, PCAE employs a locally deployed small language model (SLM) to estimate token-level surprisal scores, selectively retaining semantically critical tokens while discarding redundant ones. This significantly reduces computational overhead and transmission duration. To further enhance covert wireless transmission, we then develop a group-based proximal policy optimization (GPPO) method that samples multiple candidate actions for each state, selecting the optimal one within each group and incorporating a Kullback-Leibler (KL) divergence penalty to improve policy stability and exploration. Simulation results show that PCAE achieves comparable LLM response fidelity to baseline methods while reducing preprocessing latency by over five orders of magnitude, enabling real-time edge deployment. We further validate PCAE effectiveness across diverse LLM backbones, including DeepSeek-32B, Qwen-32B, and their smaller variants. Moreover, GPPO reduces covert transmission latency by up to 38.6% compared to existing reinforcement learning strategies, with further analysis showing that increased transmit power provides additional latency benefits.

Förderung

  • Seatrium New Energy Laboratory; in part by Singapore Ministry of Education Tier 1
  • Nanyang Technological University Centre for Computational Technologies in Finance
  • Research Innovation and Enterprise 2025 Industry Alignment Fund, Industry Collaboration Projects, administered by the Agency for Science, Technology and Resear…
  • National Natural Science Foundation of China
  • Science and Technology Development Plan Project of Jilin Province
  • China Scholarship Council
  • the China Association for Science and Technology (CAST) Young Talent Support Program for Doctoral Students

Zitieren

APA

Zhang, R., Liu, Y., Tang, S., Wang, J., Niyato, D., Sun, G., Li, Y., & Sun, S. (2026). Covert Prompt Transmission for Secure Large Language Model Services. IEEE Journal on Selected Areas in Communications, 44, 1589-1603. https://doi.org/10.1109/jsac.2025.3647412

Vancouver

Zhang R, Liu Y, Tang S, Wang J, Niyato D, Sun G, et al. Covert Prompt Transmission for Secure Large Language Model Services. IEEE Journal on Selected Areas in Communications. 2026;44:1589-1603. doi:10.1109/jsac.2025.3647412

BibTeX
@article{Zhang2026,
  author = {Zhang, Ruichen and Liu, Yinqiu and Tang, Shunpu and Wang, Jiacheng and Niyato, Dusit and Sun, Geng and Li, Yonghui and Sun, Sumei},
  title = {Covert Prompt Transmission for Secure Large Language Model Services},
  journal = {IEEE Journal on Selected Areas in Communications},
  year = {2026},
  volume = {44},
  pages = {1589--1603},
  doi = {10.1109/jsac.2025.3647412}
}
Kostenlos starten →