EUVIMEDEUVIMED entdecken
Crossref

Collaborative Knowledge Editing for Large Language Model Services in Edge-Cloud Computing

Guozeng Xu, Xiuhua Li, Junhao Wen, Qiang He, Xiaofei Wang, Victor C. M. Leung

IEEE Transactions on Services Computing · 2026-07 · 19 (4) · 2730-2745

DOI: 10.1109/tsc.2026.3703848 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Services Computing, 2026-07. Verfasst von Guozeng Xu und 5 weiteren Autor:innen. Die Arbeit nennt 54 Literaturangaben. Gefördert von: Science and Technology Innovation Key R&D Program of Chongqing, National NSFC, Natural Science Foundation of Chongqing, China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

With the rapid deployment of large language model (LLM) technology across various fields, numerous LLMs have been deployed on edge servers (ESs) to provide low-latency generative services for edge devices. However, as factual knowledge evolves, the massive number of parameters in LLMs poses significant challenges for updating LLMs on ESs. Existing studies employ federated fine-tuning to update LLMs, but these unconstrained updating approaches can lead to overfitting and knowledge forgetting, while also resulting in substantial overhead. Knowledge editing (KE), as a promising technology, ensures the injection of new knowledge while preserving existing knowledge by editing specific parameters. In this paper, we propose multi-ES collaborative KE for the first time and design the CoKE and pCoKE frameworks in edge-cloud scenarios. These frameworks enhance editing efficiency by extracting identical expressions of the same knowledge across different LLM parameters for collaborative editing across multiple ESs. Additionally, we incorporate a personalized selection module in pCoKE to provide domain-specific generative services on ESs. To further reduce editing latency, we design a binary search-based resource allocation algorithm to balance editing latency across ESs. Extensive experiments demonstrate that CoKE and pCoKE reduce editing latency by 73% while maintaining high editing quality. Moreover, pCoKE achieves an approximately 6% improvement in editing quality.

Förderung

  • Science and Technology Innovation Key R&D Program of Chongqing
  • National NSFC
  • Natural Science Foundation of Chongqing, China

Zitieren

APA

Xu, G., Li, X., Wen, J., He, Q., Wang, X., & Leung, V. C. M. (2026). Collaborative Knowledge Editing for Large Language Model Services in Edge-Cloud Computing. IEEE Transactions on Services Computing, 19(4), 2730-2745. https://doi.org/10.1109/tsc.2026.3703848

Vancouver

Xu G, Li X, Wen J, He Q, Wang X, Leung VCM. Collaborative Knowledge Editing for Large Language Model Services in Edge-Cloud Computing. IEEE Transactions on Services Computing. 2026;19(4):2730-2745. doi:10.1109/tsc.2026.3703848

BibTeX
@article{Xu2026,
  author = {Xu, Guozeng and Li, Xiuhua and Wen, Junhao and He, Qiang and Wang, Xiaofei and Leung, Victor C. M.},
  title = {Collaborative Knowledge Editing for Large Language Model Services in Edge-Cloud Computing},
  journal = {IEEE Transactions on Services Computing},
  year = {2026},
  volume = {19},
  number = {4},
  pages = {2730--2745},
  doi = {10.1109/tsc.2026.3703848}
}
Kostenlos starten →