About me
I am currently a Young Researcher at Shanghai AI Laboratory specializing in the intersection of Trustworthy Large Language Models and Intelligent Systems. My passion lies in advancing theoretical foundations and practical applications of AI, translating innovations into practical applications that ensure safe, efficient, and equitable AI applications in real-world scenarios.
I received my PhD from Zhejiang University, where I worked under the supervision of Prof. Dianhai Wang, focusing on comprehensive urban traffic state evaluation systems that have been successfully applied in Hangzhou City Brain projects. During my PhD, I worked at Imperial College London as a visiting scholar, collaborating with Prof. Washington Ochieng, FREng. I obtained my bachelor’s degree in Civil Engineering with a minor in Law from Zhejiang University.
I have authored 25+ papers at top-tier AI and interdisciplinary venues and journals including ACL, EMNLP, IEEE ITSC, and Transportation Research Part C. My interdisciplinary background allows me to bridge technological innovation with trustworthy frameworks and societal considerations, fostering solutions with real-world impact in urban systems and beyond.
📖 Educations
- 2017.09 - 2022.06, PhD, Zhejiang University, Hangzhou, China. (GPA 92.2/100, Rank 3/53)
- 2020.12 - 2021.05, Joint PhD, Imperial College London, London, UK. (Supported by CSC)
- 2013.09 - 2017.06, B.S., Zhejiang University, Hangzhou, China. (Civil Engineering, GPA 3.72/4.0)
- 2013.09 - 2017.06, Minor, Zhejiang University, Hangzhou, China. (Law, GPA 3.78/4.0)
🔥 News
- 2025.08: 🎉 Our work X-Boundary is accepted at EMNLP 2025!
- 2025.07: Our work Frontier AI Risk Management Framework in Practice: A Risk Analysis Technical Report is released!
- *2025.01: 🎉 Our work EvoBench is accepted at ACL 2025!
- 2025.01: Our survey paper on LLM watermarking for intelligence identification is published in Artificial Intelligence Review!
- 2024.06: 🎉 Our work on adaptive traffic signal control via MARL is published in Journal of Advanced Transportation!
📝 Recent Publications
Xiao Yu, Yi Yu*, Dongrui Liu, Jing Shao.
[🛡️AI Security, AIGC Detection]
Xiaoya Lu, Dongrui Liu, Yi Yu, Luxin Xu, Jing Shao
[🛡️LLM Safety, Representation Editing]
Data on the Move: Traffic-Oriented Data Trading Platform Powered by AI Agent with Common Sense
Yi Yu, Shengyue Yao, Tianchen Zhou, Yue Fu, Jingru Yu, Dianhai Wang, Xuhong Wang, Yan Chen, Yilun Lin
[🏙️ Smart Cities, LLM Application]
Building Intelligence Identification System via Large Language Model Watermarking: A Survey and Beyond, Xuhong Wang, Haoyu Jiang, Yi Yu, Jingru Yu, Yilun Lin, Pengyuan Yi, Yitong Wang, Yu Qiao, Li Li, Fei-Yue Wang, Artificial Intelligence Review, 2025, 58(8), 249. [🛡️AI Safety, LLM Watermarking]
Dive into the Agent Matrix: A Realistic Evaluation of Self-Replication Risk in LLM Agents, Boxuan Zhang, Yi Yu, Jiaxuan Guo, Jing Shao, Submitted to ICLR 2026. [🛡️AI Safety, LLM Agent]
Frontier AI Risk Management Framework in Practice: A Risk Analysis Technical Report, Shanghai AI Lab, Xiaoyang Chen, Yunhao Chen, et al. (including Yi Yu), arXiv preprint, 2025. [🛡️AI Safety, Risk Management]
SafeWork-R1: Coevolving Safety and Intelligence under the AI-45° Law, Shanghai AI Lab, Yicheng Bao, Guanxu Chen, et al. (including Yi Yu), arXiv preprint, 2025. [🛡️AI Safety, LLM Agent]
The Shadow of Fraud: The Emerging Danger of AI-Powered Social Engineering and Its Possible Cure, Jingru Yu, Yi Yu, Xuhong Wang, et al., Submitted to IEEE TSMC, 2024. [🛡️AI Safety, Social Engineering]
Improving the Urban Transport System Resilience Through Adaptive Traffic Signal Control Enabled by Decentralised Multiagent Reinforcement Learning, Xiangmin Yang, Yi Yu, Yuxiang Feng, Washington Yotto Ochieng, Journal of Advanced Transportation, 2024(1), 3035753. [🏙️ Smart Cities, MARL]
👾 Research Philosophy
“It is better to light a candle than to curse the darkness.”
I believe that the true value of AI research lies not merely in technical advancement, but in creating meaningful positive impact on society. My commitment is to develop AI systems that are not only powerful, but also responsible, equitable, and beneficial to humanity.
My interdisciplinary background enables me to approach challenges from multiple perspectives. I strive to create solutions that thoughtfully balance technological innovation with ethical considerations and societal implications. I am passionate about contributing to the academic community and nurturing the next generation of researchers, believing that through collaboration and knowledge sharing, we can collectively build AI technologies that truly serve humanity.