activity
20182025
most citedImage Shortcut Squeezing: Countering Perturbative Availability Poisons with Compression

6 citations · 26 across the 10 of their papers we have counts for

collaborators

15 papers

cs.CR2025

Typographic Attacks in a Multi-Image Setting

Xiaomeng Wang, Zhengyu Zhao, Martha Larson

Large Vision-Language Models (LVLMs) are susceptible to typographic attacks, which are misclassifications caused by an attack text that is added to an image. In this paper, we intr…

cs.CV2024★ 2 cited

Physical 3D Adversarial Attacks against Monocular Depth Estimation in Autonomous Driving

Junhao Zheng, Chenhao Lin, Jiahao Sun +3

Deep learning-based monocular depth estimation (MDE), extensively applied in autonomous driving, is known to be vulnerable to adversarial attacks. Previous physical attacks against…

cs.CV2024★ 1 cited

Adversarial Example Soups: Improving Transferability and Stealthiness for Free

Bo Yang, Hengwei Zhang, Jindong Wang +4

Transferable adversarial examples cause practical security risks since they can mislead a target model without knowing its internal knowledge. A conventional recipe for maximizing…

cs.CV2023

Collapse-Aware Triplet Decoupling for Adversarially Robust Image Retrieval

Qiwei Tian, Chenhao Lin, Zhengyu Zhao +2

Adversarial training has achieved substantial performance in defending image retrieval against adversarial examples. However, existing studies in deep metric learning (DML) still s…

cs.CR2023★ 3 cited

Robustness Over Time: Understanding Adversarial Examples' Effectiveness on Longitudinal Versions of Large Language Models

Yugeng Liu, Tianshuo Cong, Zhengyu Zhao +3

Large Language Models (LLMs) undergo continuous updates to improve user experience. However, prior research on the security and safety implications of LLMs has primarily focused on…

cs.CV2023★ 3 cited

Generative Watermarking Against Unauthorized Subject-Driven Image Synthesis

Yihan Ma, Zhengyu Zhao, Xinlei He +3

Large text-to-image models have shown remarkable performance in synthesizing high-quality images. In particular, the subject-driven model makes it possible to personalize the image…