Zhuoyan Xu

Ph.D. candidate

Department of Statistics, Computer Science
University of Wisconsin-Madison, US

Contact:
zhuoyan.xu [at] wisc [dot] edu
Office 5388, Department of Computer Sciences, UW-Madison

Google Scholar | Github | LinkedIn | CV | X (Twitter)


About Me

I am a Ph.D. candidate at the University of Wisconsin-Madison co-advised by Prof. Yingyu Liang , Prof. Yin Li and Prof. Yiqiao Zhong . I am also fortunate to work with Prof. Kris Sankaran . I obtained my M.S. degree in Computer Science from UW-Madison. Prior to that, I obtained my B.S. degree in Statistics from Wuhan University in 2019.

My research focuses on the learning and adaptation of Foundation Model, including Large Language Models and Multimodal Models, with the goal of improving their capabilities and deployment efficiency.

I am/was an Applied Scientist Intern in summer 2025/2024 at Amazon AWS AI, Bellevue. I was a Machine Learning Engineer intern in John Deere Intelligent Solutions Group in summer 2022. I also had an internship at China Merchants Bank, Shanghai.

News

Publications

* denotes equal contribution or alphabetical order.
AdaLLaVA: Learning to Inference Adaptively for Multimodal Large Language Models
Zhuoyan Xu*, Khoi Duc Nguyen*, Preeti Mukherjee, Saurabh Bagchi, Somali Chaterji, Yingyu Liang, Yin Li

ICCV 2025
[ Website ] [ Paper ] [ Code ] [ HuggingFace ]
Out-of-distribution generalization via composition: a lens through induction heads in Transformers
Jiajun Song, Zhuoyan Xu, Yiqiao Zhong

PNAS (Proceedings of the National Academy of Sciences) 2025
[ PNAS ] [ arXiv ] [ Code ] [ Blog ] [ Poster ]
TabRAG: Efficient Table Retrieval and Understanding with Large Multimodal Models
Zhuoyan Xu, Haoyang Fang, Boran Han, Bonan Min, Bernie Wang, Shuai Zhang

Work done during internship at AWS
[ abstract ]
Conv-Basis: A New Paradigm for Efficient Attention Inference and Gradient Computation in Transformers
Yingyu Liang*, Heshan Liu*, Zhenmei Shi*, Zhao Song*, Zhuoyan Xu*, Junze Yin*

arXiv, 2024
[ arXiv ]
Do Large Language Models Have Compositional Ability? An Investigation into Limitations and Scalability
Zhuoyan Xu*, Zhenmei Shi*, Yingyu Liang

COLM (Conference on Language Modeling) 2024
[ OpenReview ] [ arXiv ] [ Code ] [ Poster ]
[ Workshop ] [ Workshop Poster ] [ Workshop Slides ]
AdaInf: Adaptive Inference for Resource-Constrained Foundation Models
Zhuoyan Xu, Khoi Duc Nguyen, Preeti Mukherjee, Somali Chaterji, Yingyu Liang, Yin Li

ICML 2024 Workshop
[ OpenReview ] [ Poster ]
Why Larger Language Models Do In-context Learning Differently?
Zhenmei Shi, Junyi Wei, Zhuoyan Xu, Yingyu Liang

ICML 2024
[ Openreview ] [ arXiv ] [ Poster ]
[ Workshop ] [ Workshop Poster ]
Towards Few-Shot Adaptation of Foundation Models via Multitask Finetuning
Zhuoyan Xu, Zhenmei Shi, Junyi Wei, Fangzhou Mu, Yin Li, Yingyu Liang

ICLR 2024
[ OpenReview ] [ arXiv ] [ Code ] [ IBM Research Talk Slides ] [ Poster ] [ Video ]
[ Workshop ] [ Workshop Poster ] [ Workshop Slides ]
Spatial Transcriptomics Dimensionality Reduction using Wavelet Bases
Zhuoyan Xu, Kris Sankaran

F1000, 2022
[ F1000 Research ] [ arXiv ] [ Rpackage ] [ code ]
Generalized Tensor Regression with Covariates on Multiple Modes
Zhuoyan Xu, Jiaxin Hu, Miaoyan Wang

arXiv, 2019
[ arXiv ] [ Rpackage ]

Research & Work Experience

Research Assistant
University of Wisconsin-Madison
2022 - Now | Prof. Yingyu Liang and Prof. Yin Li
2023 - Now | Prof. Yiqiao Zhong
2021 - 2022 | Prof. Kris Sankaran
Applied Scientist Intern
Amazon AWS AI in Bellevue, WA
Summer 2025 | Debanjan Datta , Guru Nayak and Mukul Prasad
Summer 2024 | Shuai Zhang , Boran Han and Haoyang Fang
Machine Learning Engineer Intern
John Deere in Fargo, ND
Summer 2022 | Lav Thyagarrajan
Data Scientist Intern
China Merchants Bank in Shanghai, China
Summer 2018

Academic Services

Conference Reviewer at ICML 2024-2025, NeurIPS 2024, ICLR 2025, AISTATS 2025, CVPR 2025, ICCV 2025

Teaching

Teaching Assistant:
STAT 371: Introductory Applied Statistics for the Life Sciences (20 Fall)
STAT 301: Introduction to Statistical Methods (21 Spring, 21 Summer)
STAT 303-305: R for Statistics (21 Fall)
STAT 479: Statistical Data Visualization (22 Spring)

Instructor of VISP Non-credit program:
Intro to Big Data (Jan. to Feb. in 2020) [ Slides1 Slides2 ]

Misc

Prelim: [ Document Slides ]
This page has been accessed several times since Jan 20, 2024
Last updated: Aug 2, 2025