paperSeptember 2026Unreviewed

EGP-Defense: Enhancing Adversarial Robustness of LVLMs via Training-Free Edge-Guided Prompting

Bo-Yu Wang, Zi-Wen He, Xin-Jue Hu, Chi Wang, Zi-Qiang Li, Zhang-Jie Fu

ACM Transactions on Multimedia Computing, Communications, and Applications (TOMCCAP)

Abstract

Large Vision-Language Models (LVLMs) have demonstrated remarkable multimodal comprehension capabilities, achieving state-of-the-art performance across various vision-language tasks. However, their performance drops significantly when facing adversarial attacks on the visual encoder. To alleviate this issue, existing approaches often rely on adversarial training, enhancing model robustness through substantial computational cost. Unlike these methods, this paper proposes a novel, training-free adv

Framework mappings

MITRE ATLAS

AML.T0043Craft Adversarial Data

Suggested from the entry's categories.

Cite

@article{wang2026egpdefense,
  title = {{EGP-Defense: Enhancing Adversarial Robustness of LVLMs via Training-Free Edge-Guided Prompting}},
  author = {Bo-Yu Wang and Zi-Wen He and Xin-Jue Hu and Chi Wang and Zi-Qiang Li and Zhang-Jie Fu},
  year = {2026},
  month = sep,
  journal = {ACM Transactions on Multimedia Computing, Communications, and Applications (TOMCCAP)},
  doi = {10.1145/3845608},
  url = {https://www.semanticscholar.org/paper/1791b7fdd331177fbea7a13c7100d82d38bbc632}
}