Meng Chen

profile.png
Zhejiang University
Hangzhou, China

I am currently a third-year PhD student advised by Dr. Li Lu in the College of Computer Science and Technology at Zhejiang University. I also earned my bachelor’s degree from Zhejiang University in 2021.

My research interests lie in AI security with a specific focus on adversarial examples and backdoor learning, especially their practical threats to intelligent audio systems. By “debugging” deep learning, my research seeks to help build trustworthy AI systems in security-critical tasks.


News

Dec, 2023 One first-authored journal paper got accepted by TIFS!
Jul, 2023 One first-authored paper got accepted by USENIX Security 2024!
Apr, 2023 One first-authored paper got accepted by UbiComp/IMWUT!
Jan, 2023 One paper got accepted by UbiComp/IMWUT!
Oct, 2022 Our research “A Non-Intrusive and Adaptive Speaker De-Identification Scheme Using Adversarial Examples” won Best Poster Runner-up Award from MobiCom 2022!
Sep, 2022 One paper got accepted by SenSys 2022!
Aug, 2022 One first-authored poster paper got accepted in MobiCom 2022!
Mar, 2022 I received Student Conference Grant from INFOCOM 2022.
Nov, 2021 One first-authored paper got accepted in INFOCOM 2022!

Selected Publications

  1. AdvReverb: Rethinking the Stealthiness of Audio Adversarial Examples to Human Perception
    Meng ChenLi LuJiadi YuZhongjie BaFeng Lin ,  and  Kui Ren
    IEEE Trans. Inf. Forensics Secur., 2024
  2. Devil in the Room: Triggering Audio Backdoors in the Physical World
    Meng Chen ,  Xiangyu Xu ,  Li LuZhongjie BaFeng Lin ,  and  Kui Ren
    In Proceedings of USENIX Security , 2024
  3. VoiceCloak: Adversarial Example Enabled Voice De-Identification with Balanced Privacy and Utility
    Meng ChenLi Lu ,  Junhao Wang ,  Jiadi YuYingying Chen ,  Zhibo Wang ,  Zhongjie BaFeng Lin ,  and  Kui Ren
    Proc. ACM Interact. Mob. Wearable Ubiquitous Technol., 2023
  4. PhoneyTalker: An Out-of-the-Box Toolkit for Adversarial Example Attack on Speaker Recognition
    Meng ChenLi LuZhongjie Ba ,  and  Kui Ren
    In Proceedings of IEEE INFOCOM , 2022