Shengshan Hu

8

Papers

12

Total Citations

Papers (8)

Vanish into Thin Air: Cross-prompt Universal Adversarial Attacks for SAM2

Improving Generalization of Universal Adversarial Perturbation via Dynamic Maximin Optimization

Breaking Barriers in Physical-World Adversarial Examples: Improving Robustness and Transferability via Robust Feature

Detecting and Corrupting Convolution-based Unlearnable Examples

Revisiting Gradient Pruning: A Dual Realization for Defending against Gradient Attacks

Test-Time Backdoor Detection for Object Detection Models

Towards Model Extraction Attacks in GAN-Based Image Translation via Domain Shift Mitigation

NumbOD: A Spatial-Frequency Fusion Attack Against Object Detectors