Hi!
I am Zihao Xu (徐 子昊), a fourth-year PhD candidate in the Computer Science program at Rutgers University. My advisor is Professor Hao Wang, and my research focuses on Bayesian Deep Learning and Domain Adaptation. Recently, I focus on the model generalization for Large Language Models (LLM). I received my undergraduate degree in Computer Science from Shanghai Jiaotong University, a top-ranked university in China. During my time there, I was also a member of the ACM Honors Class, a prestigious program for the top 5% of Computer Science students at the university. Notable alumni of the program include Tianqi Chen (Assistant Professor at Carnegie Mellon University), Mu Li (former Senior Principal Scientist at Amazon), and Bo Li (Assistant Professor at Harvard Medical School).
News
- Sep. 27th, 2024: Our paper: Towards a Generalized Bayesian Model of Reconstructive Memory: A Generalized Model of Reconstructive Memory is accepted by Computational Brain & Behavior.
- Jun. 30th, 2023: Our paper: Taxonomy-Structured Domain Adaptation is accepted by ICML 2023. Code is released.
- Jan. 21th, 2023: Our paper: Domain-Indexing Variational Bayes: Interpretable Domain Index for Domain Adaptation is accepted by ICLR 2023 (spotlight). See our code and openreview page for more details.
- Feb. 8th, 2022: Our paper: Graph-Relational Domain Adaptation is accepted by ICLR 2022. Code is released here.
Selected Publications
“*” indicates equal contribution.
Taxonomy-Structured Domain Adaptation
Tianyi Liu* , Zihao Xu*, Hao He, Guang-Yuan Hao, Hao Wang
International Conference on Machine Learning (ICML) 2023
[paper] [code (and data)] [talk] [slides]
Domain-Indexing Variational Bayes for Domain Adaptation
Zihao Xu* , Guang-Yuan Hao*, Hao He, Hao Wang.
(Spotlight) International Conference on Learning Representations (ICLR) 2023
[paper] [code (and data)] [talk] [slides]
Graph-Relational Domain Adaptation
Zihao Xu, Hao he, Guang-He Lee, Yuyang Wang, Hao Wang
International Conference on Learning Representations (ICLR) 2022
[paper] [code (and data)] [talk] [slides] [website]