EUVIMEDEUVIMED entdecken
Crossref

Boosting Adversarial Transferability With Scaled Logit Ensemble

Jingtian Wang, Mengyao Xiao, Xiaolong Li, Jian Li, Bin Ma, Yao Zhao

IEEE Transactions on Circuits and Systems for Video Technology · 2026-11 · 36 (11) · 17589-17601

DOI: 10.1109/tcsvt.2026.3720436 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2026-11. Verfasst von Jingtian Wang und 5 weiteren Autor:innen. Die Arbeit nennt 58 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Jiangxi Province, Jinan “New 20 Universities”–Project of Introducing Innovation Team.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Adversarial transferability enables adversarial examples to attack unseen models. State-of-the-art input transformation attacks (ITAs) enhance transferability through ensemble optimization over multiple transformed inputs. However, existing ITAs mainly focus on designing more effective transformation techniques while largely overlooking the ensemble optimization strategy. As a result, most ITAs continue to rely on a loss-level ensemble mechanism, leaving substantial room for improving transferability through better ensemble optimization. To this end, a novel Scaled Logit Ensemble (SLE) optimization framework is proposed to enhance the adversarial transferability of existing ITAs. Specifically, SLE first replaces the loss-level ensemble mechanism with a logit-level ensemble strategy that has been shown to yield superior transferability in model ensemble attacks. Furthermore, to alleviate scale variations among logits generated from different transformed inputs, a Self-Standard Deviation Scaling method is proposed to adaptively calibrate the logits before aggregation. Extensive experiments demonstrate that SLE consistently improves the transferability of multiple ITAs across diverse model architectures, suggesting that ensemble optimization constitutes a promising direction for advancing ITAs.

Förderung

  • National Natural Science Foundation of China
  • Natural Science Foundation of Jiangxi Province
  • Jinan “New 20 Universities”–Project of Introducing Innovation Team
  • Natural Science Foundation of Shandong Province
  • Qilu University of Technology

Zitieren

APA

Wang, J., Xiao, M., Li, X., Li, J., Ma, B., & Zhao, Y. (2026). Boosting Adversarial Transferability With Scaled Logit Ensemble. IEEE Transactions on Circuits and Systems for Video Technology, 36(11), 17589-17601. https://doi.org/10.1109/tcsvt.2026.3720436

Vancouver

Wang J, Xiao M, Li X, Li J, Ma B, Zhao Y. Boosting Adversarial Transferability With Scaled Logit Ensemble. IEEE Transactions on Circuits and Systems for Video Technology. 2026;36(11):17589-17601. doi:10.1109/tcsvt.2026.3720436

BibTeX
@article{Wang2026,
  author = {Wang, Jingtian and Xiao, Mengyao and Li, Xiaolong and Li, Jian and Ma, Bin and Zhao, Yao},
  title = {Boosting Adversarial Transferability With Scaled Logit Ensemble},
  journal = {IEEE Transactions on Circuits and Systems for Video Technology},
  year = {2026},
  volume = {36},
  number = {11},
  pages = {17589--17601},
  doi = {10.1109/tcsvt.2026.3720436}
}
Kostenlos starten →