Zian Wang
Department of Computer Science, Stony Brook University
Contact : ziawang@cs.stonybrook.edu
About Me
I am a second year Ph.D. student studying at the Stony Brook University, Department of Computer Science. I obtained my Master’s degree in Information Science from the School of Computing and Information at the University of Pittsburgh in January 2023 and hold dual Bachelor’s degrees in Computer Science and Computing and Information Science from my undergraduate studies.
My primary research focus is on Natural Language Processing (NLP) and its Security Issues. I worked with Dr. Ting Wang in ALPS lab at Stony Brook university about AI security topics and making the AI systems free of malicious adversaries. I participated in a project about evaluating LLM text watermarking techniques and its related attacks, and helped another project about a defence against jailbreak attacks on LLMs by selectively evicting low-importance tokens from the KV cache. Now I am writing a Systematization of Knowledge (SoK) paper about security issues and related defences in LLM-based agent systems.
During my Master’s studies, I conducted research about NLP applications in the PICSO lab with Dr. Yuru Lin at UPitt. One of the endeavors is a paper proposing a predictive model for food insecurity across African nations, accepted by IEEE BigData 2023. Additionally, I finished a review project on Bayesian Optimization under the guidance of Dr. Joseph Yurko. Also, I myself have finished a paper on using Transfer Learning to apply BERT models to downstream tasks, and it has been accepted by SPML 2023.
Beyond my core research areas, I am also interested in most of the general directions in machine learning, including Trustworthy ML, Bayesian Optimization, Machine Leaning Algorithms, and Machine Learning in multi-domain integration. As a perpetual learner, I continue to navigate the vast and intriguing landscape of machine learning with enthusiasm. I am on the lookout for potential fits in research groups of our department, any opportunities or suggestions would be greatly appreciated.
Here is my CV, my transcript in UPitt , and my my current transcript in SBU.
Papers and reports
2024:
WaterPark: A Robustness Assessment of Language Model Watermarking
Jiacheng Liang, Zian Wang, Lauren Hong, Shouling Ji, Ting Wang
Submitted to Association for Computational Linguistics (ACL 2025).
RobustKV: Defending Large Language Models against Jailbreak Attacks via KV Eviction
Tanqiu Jiang, Zian Wang, Jiacheng Liang, Changjiang Li, Yuhui Wang, Ting Wang
Accepted by 2025 International Conference on Learning Representations (ICLR 2025).
2023:
HungerGist: An Interpretable Predictive Model for Food Insecurity
Yongsu Ahn, Muheng Yan, Zian Wang, and Yu-Ru Lin
Accepted by 2023 IEEE International Conference on Big Data (BigData).
2022:
A New Computationally Efficient Method to Tune BERT Networks – Transfer Learning
Zian Wang
Accepted by 2023 International Conference on Signal Processing and Machine Learning (CONF-SPML 2023).
2021:
Police Union Contract Misconduct Complaint Detection
Zian Wang, Sonal Gupta, Shuo Zheng
A final report paper for the Data Mining course, topic is given by Dr. Lin Yuru.
Thanks
I sincerely appreciate Dr. Wang’s insightful mentorship throughout our collaboration over the past one and a half years. He teaches me how to independently explore and advance in the process of conducting research, and has also helped me develop a more refined understanding and taste of high-quality academic work!
Beyond our recent collaboration, I am also deeply grateful to mentors who supported me throughout my foundational academic development.
I am deeply grateful to Dr. Lin and Dr. Yurko for their invaluable guidance during my Master’s studies and research experiences. Both professors provided immense support in lab projects, courses, and in my overall academic journey. I extend my heartfelt appreciation to them!