Graduate Student
University of Wisconsin-Madison
liu . yang @ wisc . edu
About Me
I am a PhD candidate in computer science at the University of Wisconsin–Madison. I am very fortunate to be advised by Prof. Robert D. Nowak, Prof. Dimitris Papailiopoulos and Prof. Kangwook Lee. My research focuses on the intersection of machine learning and optimization, with a recent emphasis on understanding in-context learning in controlled settings.In 2020, I earned my B.E. in Computer Science from Xi’an Jiaotong University. During my third year of undergraduate studies, I participated in an exchange program at the University of California, Berkeley. While there, I was a member of Dr. Stella Yu’s group at ICSI.
Publications
How Well Can Transformers Emulate In-context Newton's Method?Angeliki Giannou, Liu Yang, Tianhao Wang, Dimitris Papailiopoulos, Jason D. Lee
Artificial Intelligence and Statistics 2025 (AISTATS) arxiv | code
Looped Transformers are Better at Learning Learning Algorithms
Liu Yang, Kangwook Lee, Robert D. Nowak, Dimitris Papailiopoulos
International Conference on Learning Representations 2024 (ICLR) arxiv | code | 45-Minute Talk | summary
Rare Gems: Finding Lottery Tickets at Initialization
Kartik Sreenivasan*, Jy-yong Sohn*, Liu Yang, Matthew Grinde, Alliot Nagle, Hongyi Wang, Eric Xing, Kangwook Lee, Dimitris Papailiopoulos
Neural Information Processing Systems 2022 (NeurIPS) arxiv | code
A Better Way to Decay: Proximal Gradient Training Algorithms for Neural Nets
Liu Yang, Jifan Zhang, Joseph Shenouda, Dimitris Papailiopoulos, Kangwook Lee, Robert D. Nowak
Neural Information Processing Systems 2022 (NeurIPS) OPT workshop arxiv | code
Flow-based Generative Models for Learning Manifold to Manifold Mappings
Xingjian Zhen, Rudrasis Chakraborty, Liu Yang, Vikas Singh
AAAI Conference on Artificial Intelligence 2021 (AAAI) arxiv
A GMM based algorithm to generate point-cloud and its application to neuroimaging
Liu Yang, Rudrasis Chakraborty
IEEE International Symposium on Biomedical Imaging 2020 (ISBI) workshop arxiv
An "augmentation-free" rotation invariant classification scheme on point-cloud and its application to neuroimaging
Liu Yang, Rudrasis Chakraborty
IEEE International Symposium on Biomedical Imaging 2020 (ISBI) arxiv
Intrinsic Grassmann Averages for Online Linear, Robust and Nonlinear Subspace Learning
Rudrasis Chakraborty, Liu Yang, Søren Hauberg and Baba C. Vemuri
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) paper
Preprints
Task Vectors in In-Context Learning: Emergence, Formation, and BenefitsLiu Yang, Ziqian Lin, Kangwook Lee, Dimitris Papailiopoulos, Robert D. Nowak
arxiv'25 | summary
Unifying Generative and Dense Retrieval for Sequential Recommendation
Liu Yang, Fabian Paischer, Kaveh Hassani, Jiacheng Li, Shuai Shao, Zhang Gabriel Li, Yun He, Xue Feng, Nima Noorshams, Sem Park, Bo Long, Robert D Nowak, Xiaoli Gao, Hamid Eghbalzadeh
arxiv'24 | summary
Preference Discerning with LLM-Enhanced Generative Retrieval
Fabian Paischer, Liu Yang, Linfeng Liu, Shuai Shao, Kaveh Hassani, Jiacheng Li, Ricky Chen, Zhang Gabriel Li, Xialo Gao, Wei Shao, Xue Feng, Nima Noorshams, Sem Park, Bo Long, Hamid Eghbalzadeh arxiv'24 | summary
Everything Everywhere All at Once: LLMs can In-Context Learn Multiple Tasks in Superposition
Zheyang Xiong, Ziyang Cai, John Cooper, Albert Ge, Vasilis Papageorgiou, Zack Sifakis, Angeliki Giannou, Ziqian Lin, Liu Yang, Saurabh Agarwal, Grigorios G Chrysos, Samet Oymak, Kangwook Lee, Dimitris Papailiopoulos
arxiv'24 | summary
Future Prediction Can be a Strong Evidence of Good History Representation in Partially Observable Environments
Jeongyeol Kwon*, Liu Yang*, Robert Nowak, Josiah Hanna
arxiv'24
POIRot: A rotation invariant omni-directional pointnet
Liu Yang, Rudrasis Chakraborty, and Stella X. Yu
arxiv'19
Work Experiences
- Meta (2023 fall) with Minhui Huang, on exploring multi-task architectural design of the foundation model in ads ranking system.
- Meta (2024 summer) with Hamid Eghbalzadeh and Xiaoli Gao, on understanding the generative retrieval's limitation in sequential recommendation system.