Hui Wei (卫慧)
Resume: PDF (updated 6/2025)
Interests: Trustworthy Intelligence, AI Safety, Computer Vision
Contact: weihui.1308@gmail.com, hui.wei@oulu.fi
Links: DBLP · Google Scholar · Github
Hui Wei is currently a Postdoctoral Researcher at the University of Oulu, Finland, working under the supervision of Prof. Guoying Zhao. Before this, he earned his Ph.D. in Computer Science from Wuhan University in 2025, supervised by Prof. Zheng Wang. His research focuses on AI Safety, with an emphasis on the trustworthiness of AI systems deployed in real-world scenarios.
His motto is: Doing Honest Work with My Own Hands.
News
- [2025.08.04] Two co-authored papers on physical adversarial attack and jailbreak attack are accepted by ACM MM 2025.
- [2025.03.06] One paper on physical adversarial attack is accepted to CVPR 2025.
- [2024.12.31] One co-authored paper on privacy protection is accepted to AAAI 2025.
- [2024.09.26] One paper on physical adversarial attack is accepted to NeurIPS 2024. The source code is released. Feel free to contact me for questions.
- [2024.07.13] Our survey paper on physical adversarial attack is accepted by T-PAMI.
- [2023.11.01] I've been awarded the Second Prize of Academic Innovation of Wuhan University.
- [2023.10.28] We won the first prize of the 18th China "Challenge Cup" competition.
- [2023.10.13] I've been awarded 2023 China National Scholarship, Wuhan University.
- [2023.09.19] One paper on physical backdoor attack is accepted to ACM MM 2023. The source code is released.
- [2023.01.08] We won the first prize in the "Huawei Cup" Competition.
- [2022.12.06] One paper on physical adversarial attack is accepted to AAAI 2023. The source code is released.
Last Updated: Sept. 04, 2025