Fudan MAS

Artificial Intelligence

The group is interested in the study of novel methods in AI and their security issues. The key researches are two-folded: 1) AI security, including adversarial examples, neural network backdoors, neural network model watermarking, etc. 2) chatbot in social media based on AI technology.

learning resource

Neural Network Watermarking

publication

  1. Forward Creation, Reverse Selection: Achieving Highly Pertinent Multimodal Responses in Dialogue Context
    Ge Luo, Manman Zhang，Yuchen Ma，Sheng Li, Zhenxing Qian*, and Xinpeng Zhang*
    Proceedings of the 32st ACM International Conference on Information & Knowledge Management, 2023.
  2. Physical Invisible Backdoor Based on Camera Imaging
    Y. Guo, N. Zhong, Z. Qian, and X. Zhang
    MM '23: Proceedings of the 30th ACM International Conference on Multimedia, 2023.
  3. ACM MM
    Deep Neural Network Watermarking against Model Extraction Attack
    J. Tan, N. Zhong, Z. Qian, X. Zhang, and Li
    MM '23: Proceedings of the 30th ACM International Conference on Multimedia, 2023.
    Website
  4. ACM MM
    VCMaster: Generating Diverse and Fluent Live Video Comments Based on Multimodal Contexts
    M. Zhang, G. Luo, Y. Ma, S. Li, Z. Qian, and X. Zhang
    MM '23: Proceedings of the 30th ACM International Conference on Multimedia, 2023.
    Website
  5. Federated Learning with Label-Masking Distillation
    J. Lu, S. Li, K. Bao, P. Wang, Z. Qian, and S. Ge
    MM '23: Proceedings of the 30th ACM International Conference on Multimedia, 2023.
  6. Watermarks for Generative Adversarial Network Based on Steganographic Invisible Backdoor
    Y. Zeng, J. Tan, Z. You, Z. Qian, and X. Zhang
    IEEE International Conference on Multimedia and Expo (ICME), 2023
  7. 抑制图像非语义信息的通用后门防御策略
    郭钰生, 钱振兴, 张新鹏, 柴洪峰
    中国图象图形学报, 录用
  8. Removing Watermarks for Image Processing Networks Via Referenced Subspace Attention
    Y. Xue, Y. Zhu, Z. Zhu, S. Li, Z. Qian, and X. Zhang
    The Computer Journal, 2022.
  9. Steganography of Steganographic Networks
    G. Li, S. Li, M. Li, X. Zhang, and Z. Qian
    Thirty-Seventh AAAI Conference on Artificial Intelligence (AAAI-23), 2023
  10. Imperceptible Backdoor Attack: From Input Space to Feature Representation
    N. Zhong, Z. Qian, and X. Zhang
    the 31st International Joint Conference on Artificial Intelligence (IJCAI 22), pp. 1736-1742, July 23~July 29, Messe Wien, Vienna, Austria, 2022
  11. HF-Defend: Defending Against Adversarial Examples Based on Halftoning
    G. Liu, S. Li, Z. Qian, and X. Zhang
    IEEE 24th International Workshop on Multimedia Signal Processing, September 26-28, 2022, Shanghai, China.
  12. Privacy-Preserving Student Learning with Differentially Private Data-Free Distillation
    B. Liu, J. Lu, P. Wang, J. Zhang, D. Zeng, Z. Qian, and S. Ge
    IEEE 24th International Workshop on Multimedia Signal Processing, September 26-28, 2022, Shanghai, China.
  13. Robust Backdoor Injection with the Capability of Resisting Network Transfer
    L. Feng, S. Li, Z. Qian, and X. Zhang
    Information Science, 612(2022): 594-611.
  14. Perceptual Hash of Neural Networks
    Z. Zhu, H. Zhou, S. Xing, Z. Qian, S. Li, and X. Zhang
    Symmetry, accepted.
  15. Invertible Image Dataset Protection
    X. Zeng, K. Chen, Q. Ying, S. Li, Z. Qian, and X. Zhang
    IEEE International Conference on Multimedia and Expo (ICME) 2022, Taipei, Taiwan, 18-22 July, 2022
  16. Unlabeled Backdoor Poisoning in Semi-Supervised Learning
    L. Feng, S. Li, Z. Qian, and X. Zhang
    IEEE International Conference on Multimedia and Expo (ICME) 2022, Taipei, Taiwan, 18-22 July, 2022
  17. Encryption Resistant Deep Neural Network Watermarking
    G. Li, S. Li, Z. Qian, and X. Zhang
    2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2022), 22-27 May 2022, Singapore, accepted.
  18. Object-Oriented Backdoor Attack Against Image Captioning
    M. Li, N. Zhong, X. Zhang, Z. Qian, and Sheng Li
    2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2022), 22-27 May 2022, Singapore, accepted.
  19. Stealthy Backdoor Attack with Adversarial Training
    L. Feng, S. Li, Z. Qian, and X. Zhang
    2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2022), 22-27 May 2022, Singapore, accepted.
  20. Joint Learning for Addressee Selection and Response Generation in Multi-Party Conversation
    Q. Song, S. Li, P. Wei, L. Ge, X. Zhang, and Z. Qian
    2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2022), 22-27 May 2022, Singapore, accepted.
  21. Deep Neural Network Retrieval
    N. Zhong, Z. Qian, and X. Zhang
    MM '21: Proceedings of the 29th ACM International Conference on Multimedia, October 2021, pp 3455-3463
  22. Fragile Neural Network Watermarking with Trigger Image Set
    R. Zhu, P. Wei, S. Li, Z. Yin, X. Zhang and Z. Qian
    International Conference on Knowledge Science, Engineering and Management (KSEM 2021), Tokyo, Japan, 2021
  23. Undetectable Adversarial Examples based on Microscopical Regularization
    N. Zhong, Z. Qian, and X. Zhang
    IEEE International Conference on Multimedia and Expo (ICME) 2021, Shenzhen, China, 5-9 July 2021
  24. On Generating JPEG Adversarial Images
    M. Shi, S. Li, Z. Yin, X. Zhang, and Z. Qian
    IEEE International Conference on Multimedia and Expo (ICME) 2021, Shenzhen, China, 5-9 July 2021
  25. Reversible Privacy-Preserving Recognition
    Z. You, S. Li, Z. Qian, and X. Zhang
    IEEE International Conference on Multimedia and Expo (ICME) 2021, Shenzhen, China, 5-9 July 2021
  26. Key based Artificial Fingerprint Generation for Privacy Protection
    S. Li, X. Zhang, Z. Qian, G. Feng, and Y. Ren
    IEEE Transactions on Dependable and Secure Computing, 2018, DOI: 10.1109/TDSC.2018.2812192
  27. Dynamic adjustment of hidden node parameters for extreme learning machine
    G. Feng, Y. Lan, X. Zhang, and Z. Qian.
    IEEE Transactions on Cybernetics, 45(2): 279-288，2015.
  28. Evolutionary selection extreme learning machine optimization for regression
    G. Feng, Z. Qian, and X. Zhang
    Soft Computing, 16: 1485-1491, 2012.