About me

Hi! This is Minxing Zhang (张敏行), and I’m a second-year PhD student in CISPA supervised by Michael Backes and Xiao Zhang. I obtained my B.S. degree in Computer Science and Technology from Shandong University (2020) advised by Zhaochun Ren.

My CV is available here.

Happy that I have worked with many intelligent researchers. Looking forward to future collaborations on interesting projects.

Research Interests

• Trustworthy Machine Learning
• AI Security
• Data Privacy

What’s New!

• [10/2024] Our paper titled Generating Less Certain Adversarial Examples Improves Robust Generalization is accepted by TMLR.
• [10/2024] Our paper titled Invisibility Cloak: Disappearance under Human Pose Estimation via Backdoor Attacks is available online, which proposed the first disappearance attack against HPE via backdoor techniques.
• [08/2024] My implementations of PGD-based adversarial training are available online.
• [08/2024] Our paper titled Vera Verto: Multimodal Hijacking Attack is available online, which hijacks a CV victim model to implement the adversary’s own NLP task with stealthiness.
• [10/2023] Our paper titled Generated Distributions Are All You Need for Membership Inference Attacks Against Generative Models is accepted by IEEE/CVF WACV2024.
• [10/2023] Our paper titled Generating Less Certain Adversarial Examples Improves Robust Generalization is available online, which proposes to improve robust generalization by our novel definition Adversarial Certainty.
• [10/2022] Start my PhD study at CISPA.
• [09/2021] Our paper titled Membership Inference Attacks Against Recommender Systems is accepted by ACM CCS 2021.
• [05/2021] Join CISPA as a preparatory-phase student.
• [06/2020] Obtain my bachelor’s degree from Shandong University.