Junyi Zhu

发表

BadDet: Backdoor Attacks on Object Detection

Xiaolu Zhang, Yinpeng Dong, Shih-Han Chan, 2022, ECCV Workshops.

Robustness and Accuracy Could Be Reconcilable by (Proper) Definition

Min Lin, Shuicheng Yan, Tianyu Pang, 2022, ICML.

Improving transferability of 3D adversarial attacks with scale and shear transformations

Yinpeng Dong, Junyi Zhu, Jihong Zhu, 2022, 2211.01093.

Towards generalizable detection of face forgery via self-guided model-agnostic learning

X. Yang, Hang Su, Yinpeng Dong, 2022, Pattern Recognit. Lett..

ViewFool: Evaluating the Robustness of Visual Recognition to Adversarial Viewpoints

Xingxing Wei, Hang Su, Yinpeng Dong, 2022, NeurIPS.

Isometric 3D Adversarial Examples in the Physical World

Yinpeng Dong, Xiao-Shan Gao, Junyi Zhu, 2022, NeurIPS.

A Comprehensive Study on Robustness of Image Classification Models: Benchmarking and Rethinking

X. Yang, Hang Su, Yuan He, 2023, ArXiv.

R-GAP: Recursive Gradient Attack on Privacy pdf

Matthew B. Blaschko, Matthew Blaschko, Junyi Zhu, 2020, ICLR.

Controllable Evaluation and Generation of Physical Adversarial Patch on Face Recognition

X. Yang, Hang Su, Tianyu Pang, 2022, ArXiv.

Rethinking Model Ensemble in Transfer-based Adversarial Attacks

Yinpeng Dong, Junyi Zhu, Huanran Chen, 2023, ArXiv.

Learning Sample Difficulty from Pre-trained Models for Reliable Prediction

Zhijie Deng, Yinpeng Dong, Junyi Zhu, 2023, ArXiv.

Surrogate Model Extension (SME): A Fast and Accurate Weight Update Attack on Federated Learning

Matthew B. Blaschko, Junyi Zhu, Ruicong Yao, 2023, ICML.

Differentially Private SGD with Sparse Gradients pdf

Matthew B. Blaschko, Matthew Blaschko, Junyi Zhu, 2021, ArXiv.

Improving Differentially Private SGD via Randomly Sparsified Gradients

Matthew B. Blaschko, Junyi Zhu, 2021, 2112.00845.

Confidence-Aware Personalized Federated Learning via Variational Expectation Maximization

Matthew B. Blaschko, Junyi Zhu, Xingchen Ma, 2023, 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).