- I am a first-year MPhil student in the Department of Artificial Intelligence at The Hong Kong University of Science and Technology, supervised by Prof. Hao LIU. HKUST ranked #29 globally in Computer Science and #1 in Hong Kong.
- Prior to my MPhil, I received a Bachelor’s degree from the Department of Electronic Engineering at Tsinghua University, a university that ranked #12 globally and #1 in Electrical and Electronic Engineering.
- My research interests are focused on foundation models, graph learning, and spatio-temporal data mining. Specifically, my current work revolves around the development of reliable, trustworthy Large Models (LLMs, LVMs) and their applications in real-world scenarios (e.g., smart cities, graph reasoning).
- I warmly welcome everyone to discuss academic matters with me. Feel free to contact me via WeChat at Bill2998 or email at [email protected]. Looking forward to your valuable insights and suggestions!
Acknowledgment
- I used to suffer from severe lumbar and heart disease from 2020 to 2022. Despite facing severe health issues, I persevered with the support of my family, friends, and supervisors, including Prof. Yong Li and Prof. Minghua Chen, as well as Dr. Junbo Zhang, Dr. Jie Feng, Dr. Yuanwei Fang, and Kai Zhao. I am deeply grateful for their help.
- After recovery, I am now determined and motivated to move on and make the best of each day.
Genius only means hard-working all one's life. (Mendeleev)
🔥 News
- 2024.02: We designed LightGPT to improve the LLMLight framework, using imitation fine-tuning from GPT-4 and policy refinement with a critic model for enhanced, cost-effective control policies in diverse traffic scenarios. Find more details HERE.
- 2024.01: We have conducted a comprehensive review of Urban Foundation Models, offering insightful perspectives and findings. Access our detailed paper and repository HERE.
- 2023.12: Our latest work applies Large Language Models (LLMs) to Traffic Signal Control (TSC) tasks. Discover our paper, code, and demo HERE. We are currently preparing further studies for submission to KDD 2024.
- 2023.08: I accepted the MPhil offer from HKUST, and it was my pleasure to join USAIL group.
- 2023.08: We went to Beijing to participate in The First Research Summit of Urban Science. Our poster was accepted by Tsinghua, and I gave a quick presentation in Urban Cup.
- 2023.05: 🎉 I joined the USAIL Lab of AI thrust as a research assistant at The Hong Kong University of Science and Technology
. My supervisor is Prof. Hao Liu, and our group works perfectly harmoniously. - 2022.12: 🎉🎉 I recovered from severe Spinal Disc Herniation thanks to rehabilitation training and swimming!
Older News
- 2021.07: I received the algorithm intern offer from the Machine Learning Platform of KuaiShou Technology.
- 2021.01: I joined Alibaba DAMO Academy
as a research intern!
📖 Educations
- 2025 (Anticipated), The Hong Kong University of Science and Technology
Master of Philosophy, Department of Artificial Intelligence
Supervisor: Prof. Hao Liu Co-supervisor: Prof. Hui Xiong, Prof. Xiaowen Chu - 2022, Tsinghua University
Bachelor, Department of Electronic Engineering
Supervisor: Prof. Yong Li - 2019.07 - 2019.10, Pennsylvania State University
Undergraduate Visiting Researcher, School of Electronic Engineering and Computer Science
📝 Projects and Publications
-
Ongoing“Defend Against Adaptive Jailbreak Attack for LLMs”
Targeting NeurIPS 2024 -
Ongoing“Corrective Large Language Model Unlearning”
Targeting NeurIPS 2024
- “Towards Urban General Intelligence: A Review and Outlook of Urban Foundation Models”
[Github], [Press (in Chinese)]
Weijia Zhang, Jindong Han, Zhao Xu, Hang Ni, Hao Liu, Hui Xiong
arXiv:2402.01749 [cs.CY]
- “Large Language Models as Traffic Signal Control Agents: Capacity and Opportunity”
[Github], [Project Page], [Demo]
Siqi Lai, Zhao Xu, Weijia Zhang, Hao Liu, Hui Xiong
arXiv:2312.16044 [cs.AI] Under Review of KDD 2024 ADS Track.
-
TMC 2020“DeepMM: Deep learning based map matching with data augmentation” (Extension of SIGSPATIAL 2019 Paper)
Jie Feng, Yong Li, Kai Zhao, Zhao Xu, Tong Xia, Lin Chen, Depeng Jin
IEEE Transactions on Mobile Computing (CCF A) -
SIGSPATIAL 2019“DeepMM: Deep learning based map matching with data augmentation”
Kai Zhao, Jie Feng, Zhao Xu, Tong Xia, Lin Chen, Funing Sun, Diansheng Guo, Depeng Jin, Yong Li
Proceedings of the 27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems (CORE Rank A, Acceptance Rate: 21%)
🍀 Services
- Reviewer: LLMAgents@ICLR2024
🎖 Honors and Awards
- Scholarship for Outstanding Freshman (Top 1%, 30 out of 3000)
- Outstanding Student in Work-Study Assistance
💻 Work Experience
- 2023.05 - Present, Research Assistant, AI Thrust, The Hong Kong University of Science and Technology .
- 2021.01 - 2021.05, Research Intern, Computing Technology Lab, Alibaba DAMO Academy .
- 2018.05 - 2020.10, Research Assistant, Future Intelligence laB (FIB), Tsinghua University
. - 2019.07 - 2019.10, Research Assistant, School of Electronic Engineering and Computer Science
.
💬 To Prospective Advisor
Details
English Version
I plan to apply for Ph.D. programs for Fall 2025. Here are a few expectations I have for my future advisor and collaborators:
- Interest: My motivation is fueled by my passions, and working on projects that don’t spark my interest is a challenge I prefer to avoid. It’s crucial for me to find a research group where our interests align, as this is a fundamental requirement for me.
- Qualities: I strive to uphold values of integrity, diligence, and responsibility, and I seek to work alongside individuals who share these commendable traits.
- Atmosphere: A positive team environment significantly boosts my motivation and productivity. This is a key reason why I chose to join USAIL. The importance of the above factors is ranked as follows: 1 = 2 > 3 > > others.
Recently, I wrote a blog in Chinese, detailing my expectations for Ph.D. applications, future research, and life. You’re welcome to read it: 24#00 Before Applying
Thank you for taking the time to learn more about me. If you have any feedback or suggestions, please don’t hesitate to reach out to me via email ([email protected]), phone (+86-13260261866), or WeChat (Bill2998).
Chinese Version
我计划申请 25Fall 入学的 Ph.D. 项目，以下几点是我对未来的导师与合作者的期待：
- 兴趣：我一直是一个兴趣导向的人，我无法忍受做完全不感兴趣的事情，相近的研究兴趣是必要条件。
- 品质：我以诚信、勤奋、责任心等准则要求自己，所以我也希望能与具有类似优秀品质的人合作。
- 氛围：在拥有一个良好氛围的团队中，我会有更高的自我驱动和执行力 (这是我加入 USAIL 的一个重要原因)。 以上几点的重要程度：1 = 2 > 3 > > 其他
最近我写了一篇中文博客，更详细地介绍了我对 PhD 申请、未来的研究与生活的期待，欢迎阅读：24#00 写在申请前
最后，感谢您愿意花时间来了解我，如果您对我有任何意见和建议，欢迎通过邮件 ([email protected])、电话 (+86-13260261866) 或微信 (Bill2998) 联系我。
Know More About Me in 1 Minute
- 我一直是一个兴趣导向的人，我无法忍受做完全不感兴趣（例如模拟电路）的事情，同时我对新知识和新领域的挑战非常感兴趣且具备足够的行动力（例如，尽管我不喜欢模拟电路，但我对芯片设计很感兴趣，所以在本科期间我熟练 Verilog 语言后，自己选修了微电子的研究生课程，了解并实践了芯片设计的完整流程，完成了一个简单卷积神经网络加速芯片的版图设计）。
- 我患有较为严重的腰椎间盘突出疾病，并曾因病影响了两年的学习时间。尽管腰椎疾病是不可逆的，通过坚持康复训练和定期游泳等方式，目前疾病已经不会影响到我正常的生活和工作。
- 生病期间，我在病痛中花了很长时间思考对人生的规划，确定了真正的兴趣，对未来的期待是我战胜疾病的一个重要原因。这也使我决定将有限的时间投入到感兴趣的领域，也是我决定读 PhD 的原因。我是一个终身学习者，也是“日拱一卒”想法的践行者。尽管由于之前生病耽误了很多时间，但我从康复以来一直（努力:）保持对时间的高效利用和对自身的有效管理，我有信心在接下来几年内赶上并超过先行者。
last update: March 8, 2024