avatar

Yang Bai

Senior Research Engineer
ByteDance
baiyang0522 (at) gmail.com


About Me

I am currently a senoir research engineer at ByteDance. I received my Ph.D. from Tsinghua Univeristy in June 2022, where I was fortunate to be advised by Prof. Yong Jiang and Prof. Shu-tao Xia. I was also co-supervised by Prof. Yisen Wang. Prior to this, I obtained my bachelor’s degree from Tsinghua University in June 2017. I joined ByteDance in Aug 2024. Before that, I has worked in Tencent Zhuque Lab since 2022 and interned in Tencent AI Lab in 2020.

My general research interest is Responsbile GenAI – with a special focus on building GenAI systems which are data-efficient and robust, especially for commercial foundation models.

News

Publications [Google Scholar][DBLP]

  1. Special Characters Attack: Toward Scalable Training Data Extraction From Large Language Models
    Yang Bai, Ge Pei, Jindong Gu, Yong Yang#, Xingjun Ma
    (# Corresponding authors)
    PDF

  2. Cheating Suffix: Targeted Attack to Text-To-Image Diffusion Models with Multi-Modal Priors
    Dingcheng Yang, Yang Bai#, Xiaojun Jia, Yang Liu, Xiaochun Cao, Wenjian Yu#
    (# Corresponding authors)
    PDF Code

  3. Inducing High Energy-Latency of Large Vision-Language Models with Verbose Images
    Kuofeng Gao, Yang Bai, Jindong Gu, Shu-Tao Xia, Philip Torr, Zhifeng Li#, Wei Liu#
    (# Corresponding authors)
    ICLR 2024
    PDF Code

  4. OT-Attack: Enhancing Adversarial Transferability of Vision-Language Models via Optimal Transport Optimization
    Dongchen Han*, Xiaojun Jia*, Yang Bai, Jindong Gu, Yang Liu, Xiaochun Cao
    (* Equal contributions)
    PDF

  5. Distributed Adaptive Optimization with Divisible Communication
    An Xv, Yang Bai
    ECML KDD 2023
    PDF

  6. Cross Model Parallelism for Faster Bidirectional Training of Large Convolutional Neural Networks
    An Xv, Yang Bai
    ECML KDD 2023
    PDF

  7. BackdoorBox: A python toolbox for backdoor learning
    Yiming Li*, Mengxi Ya*, Yang Bai, Yong Jiang, Shu-Tao Xia
    (* Equal contributions)
    ICLR 2023 Workshop on Backdoor Attacks and Defenses in Machine Learning
    PDF Code

  8. Backdoor Defense via Adaptively Splitting Poisoned Dataset
    Kuofeng Gao*, Yang Bai*, Jindong Gu, Yong Yang, Shu-Tao Xia
    (* Equal contributions)
    CVPR 2023
    PDF Code

  9. Query efficient black-box adversarial attack on deep neural networks
    Yang Bai*, Yisen Wang*, Yuyuan Zeng, Yong Jiang, Shu-Tao Xia
    (* Equal contributions)
    Pattern Recognition
    PDF

  10. Untargeted backdoor watermark: Towards harmless and stealthy dataset copyright protection
    Yiming Li*, Yang Bai*, Yong Jiang, Yong Yang, Shu-Tao Xia, Bo Li
    (* Equal contributions)
    NeurIPS 2022
    PDF

  11. The Robust and Harmless Model Watermarking
    Yiming Li, Linghui Zhu, Yang Bai, Yong Jiang, Shu-Tao Xia
    Digital Watermarking for Machine Learning Model: Techniques, Protocols and Applications
    PDF

  12. Watermark vaccine: Adversarial attacks to prevent watermark removal
    Xinwei Liu, Jian Liu, Yang Bai, Jindong Gu, Tao Chen, Xiaojun Jia#, Xiaochun Cao#
    (# Corresponding authors)
    ECCV 2022
    PDF Code

  13. MOVE: Effective and Harmless Ownership Verification via Embedded External Features
    Yiming Li, Linghui Zhu, Xiaojun Jia, Yang Bai, Yong Jiang, Shu-Tao Xia, Xiaochun Cao
    PDF Code

  14. Test-time Adaptation of Residual Blocks against Poisoning and Backdoor Attacks
    Arnav Gudibande, Xinyun Chen, Yang Bai, Jason Xiong, Dawn Song
    CVPR 2022 Workshop
    PDF

  15. D2Defend: Dual-Domain based Defense against Adversarial Examples
    Xin Yan, Yanjie Li, Tao Dai, Yang Bai, Shu-Tao Xia
    IJCNN 2021
    PDF

  16. Clustering Effect of (Linearized) Adversarial Robust Models
    Yang Bai, Xin Yan, Yong Jiang, Shu-Tao Xia, Yisen Wang
    NeurIPS 2021
    PDF Code

  17. Improving Adversarial Robustness via Channel-wise Activation Suppressing
    Yang Bai*, Yuyuan Zeng*, Yong Jiang, Shu-Tao Xia, Xingjun Ma, Yisen Wang
    (* Equal contributions)
    ICLR 2021
    PDF Code

  18. Adaptive Frequency Learning in Two-branch Face Forgery Detection
    Neng Wang, Yang Bai, Kun Yu, Yong Jiang, Shu-tao Xia, Yan Wang
    PDF

  19. Improving Query Efficiency of Black-box Adversarial Attack
    Yang Bai*, Yuyuan Zeng*, Yong Jiang, Yisen Wang, Shu-Tao Xia, Weiwei Guo
    (* Equal contributions)
    ECCV 2020
    PDF Code

  20. Self-adaptive Feature Fool
    Xinyi Liu, Yang Bai, Shu-Tao Xia, Yong Jiang
    ICASSP 2020
    PDF

  21. Hilbert-based Generative Defense for Adversarial Examples
    Yang Bai*, Yan Feng*, Yisen Wang, Tao Dai, Shu-Tao Xia, Yong Jiang
    (* Equal contributions)
    ICCV 2019
    PDF

Writings

Services

Area Chairs

Conference Reviewers

Journal Reviewers

Contact

Address: 1199 Coleman Avenue, San Jose, CA 95110
Office Location: ByteDance Inc.
Email: baiyang0522 (at) gmail.com