Dr Zihan Wang
Postdoctoral Research Fellow
School of Electrical Engineering and Computer Science
Journal Articles
Wan, Liuhuo, Yan, Chuan, Wang, Zihan, Meng, Mark Huasong, Wang, Kailong, Wang, Haoyu, Bai, Guangdong and Dong, Jin Song (2026). Beyond the scope: Security testing of permission management in team workspaces. IEEE Transactions on Dependable and Secure Computing, 23 (5), 11701-11715. doi: 10.1109/tdsc.2026.3715679
Wang, Haodi, Wang, Zihan, Dong, Kai, Wang, Jiakai, Liu, Xianglong and Bai, Guangdong (2026). Leveraging robustness-aware channel activation for privacy protection and tracing forensics. IEEE Transactions on Dependable and Secure Computing, 23 (4), 8745-8759. doi: 10.1109/tdsc.2026.3688154
Ma, Zhongkui, Wang, Zihan and Bai, Guangdong (2025). Convex hull approximation for activation functions. Proceedings of the ACM on Programming Languages, 9 (OOPSLA2) 308, 1007-1033. doi: 10.1145/3763086
Wang, Zihan, Byrnes, Olivia, Wang, Hu, Sun, Ruoxi, Ma, Congbo, Chen, Huaming, Wu, Qi and Xue, Minhui (2023). Data hiding with deep learning: a survey unifying digital watermarking and steganography. IEEE Transactions on Computational Social Systems, 10 (6), 1-15. doi: 10.1109/tcss.2023.3268950
Conference Papers
Wang, Chi, Wang, Hanwen, Xia, Yu, Wang, Zihan and Bai, Guangdong (2027). Caliber: cross-architecture extraction-cost control for score-returning APIs. 20th International Conference, ProvSec 2026, Queenstown, New Zealand, 24-26 September 2026. Heidelberg, Germany: Springer. doi: 10.1007/978-981-92-5737-9_3
Wang, Zihan, Ma, Ethan, Ma, Zhongkui, Liu, Shuofeng, Liu, Akide, Wang, Derui, Xue, Minhui and Bai, Guangdong (2026). Catch-only-one: non-transferable examples for model-specific authorization. Neural Information Processing Systems (NeurIPS), Sydney, Australia, 6-10 December 2026.
Terashima, Shogo, Singh, Priyanka, Wang, Zihan and Bai, Guangdong (2026). ExactUMM: Exact task and data Unlearning via Model Merging. The 1st International ACM Workshop on Impact-aware Multimodal Persuasive Analysis and Contextual Trust (IMPACT '26), Rio de Janeiro, Brazil, 10 November, 2026. New York, NY, United States: ACM. doi: 10.1145/3841453.3841499
Wang, Zihan, Ma, Zhongkui, Feng, Xinguo, Yan, Chuan, Liu, Dongge, Sun, Ruoxi, Wang, Derui, Xue, Minhui and Bai, Guangdong (2026). Re-key-free, risky-free: adaptable model usage control. 2026 IEEE 11th European Symposium on Security and Privacy (EuroS&P), Lisbon, Portugal, 6 - 10 July 2026. Piscataway, NJ, United States: Institute of Electrical and Electronics Engineers. doi: 10.1109/eurosp68448.2026.00051
Guo, Zinan, Wang, Zihan, Yan, Chuan, Wan, Liuhuo, Ma, Ethan and Bai, Guangdong (2026). Variational feature compression for model-specific representations. International Workshop on Security in Machine Learning and its Applications (SiMLA), Stony Brook, NY, United States, 22–25 June 2026.
Feng, Xinguo, Ma, Zhongkui, Wang, Zihan, Abuadbba, Alsharif and Bai, Guangdong (2026). Mitigating gradient inversion risks in language models via token obfuscation. 21st Asia Conference on Computer and Communications Security-ASIACCS-Annual, Bengaluru, India, 01-05 June, 2026. New York, NY, United States: ACM. doi: 10.1145/3779208.3785389
Wang, Zihan, Ma, Zhongkui, Feng, Xinguo, Mei, Zhiyang, Ma, Ethan, Wang, Derui, Xue, Minhui and Bai, Guangdong (2025). AI model modulation with logits redistribution. 2025 Web Conference-WWW, Sydney, NSW, Australia, 28 April - 2 May 2025. New York, NY, United States: ACM. doi: 10.1145/3696410.3714737
Feng, Xinguo, Ma, Zhongkui, Wang, Zihan, Chegne, Eu Joe, Ma, Mengyao, Abuadbba, Alsharif and Bai, Guangdong (2024). Uncovering Gradient Inversion Risks in Practical Language Model Training. 31st Conference on Computer and Communications Security, Salt Lake City, UT United States, 14-18 October 2024. New York, NY United States: Association for Computing Machinery. doi: 10.1145/3658644.3690292
Wang, Zihan, Ma, Zhongkui, Feng, Xinguo, Sun, Ruoxi, Wang, Hu, Xue, Minhui and Bai, Guangdong (2024). CORELOCKER: neuron-level usage control. 2024 IEEE Symposium on Security and Privacy (SP), San Francisco, CA, United States, 19-23 May 2024. Piscataway, NJ, United States: IEEE. doi: 10.1109/sp54263.2024.00233
Liu, Shuofeng, Wang, Zihan, Xue, Minhui, Wang, Long, Zhang, Yuanchao and Bai, Guangdong (2024). Being Transparent is Merely the Beginning: Enforcing Purpose Limitation with Polynomial Approximation. 33rd USENIX Security Symposium, Philadelphia, PA United States, 14-16 August 2024. Berkeley, CA United States: USENIX Association.
Liu, Liyang, Wang, Zihan, Phan, Minh Hieu, Zhang, Bowen, Ge, Jinchao and Liu, Yifan (2024). BPKD: boundary privileged knowledge distillation for semantic segmentation. 2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), Waikoloa, HI, United States, 3 - 8 January 2024. Piscataway, NJ, United States: Institute of Electrical and Electronics Engineers. doi: 10.1109/WACV57701.2024.00110
Ma, Zhongkui, Feng, Xinguo, Wang, Zihan, Liu, Shuofeng, Ma, Mengyao, Guan, Hao and Meng, Mark Huasong (2023). Formalizing robustness against character-level perturbations for neural network language models. 24th International Conference on Formal Engineering Methods, Brisbane, QLD Australia, 21-24 November 2023. Singapore, Singapore: Springer Nature Singapore. doi: 10.1007/978-981-99-7584-6_7
Hu, Pingyi, Wang, Zihan, Sun, Ruoxi, Wang, Hu and Xue, Minhui (2022). M^4I: multi-modal models membership inference. Advances in Neural Information Processing Systems 35, New Orleans, Louisiana, United States, 28 November - 9 December 2022. San Diego, California, USA: Neural Information Processing Systems Foundation, Inc. (NeurIPS). doi: 10.52202/068431-0136
Thesis
Wang, Zihan (2026). AI usage control. PhD Thesis, School of Electrical Engineering and Computer Science, The University of Queensland. doi: 10.14264/1da3166