About

I am a fourth year Ph.D. candidate at the University of Georgia, now working with Prof. Ninghao Liu. I obtained my bachelor’s degree in statistics from the School of Statistics and Information at Shanghai University of International Business and Economics (SUIBE), where I worked with Prof. Chengcheng Hao and Prof. Liangliang Liu.

I am actively looking for a (Applied) Research Scientist position starting in December 2025.

Research Interests

Large Language Model: mechanism interpretation, evaluation, and human alignment.

Application of LLMs: recommender systems and science education.

News

  • Feb, 2025: One paper is accepted by TKNL!
  • Jan, 2025: One paper is accepted by NAACL 2025!
  • Nov, 2024: I passed my comprehensive exam!
  • Nov, 2024: I finished my first half-marathon!
  • Jul, 2024: One paper is accepted by CIKM 2024!
  • Jun, 2024: I joined the Samsung Ads (@San Jose) as a research intern!
  • Apr, 2024: One paper is accepted by ACL 2024!
  • Apr, 2024: Invited talk in the School of Computing at the University of Georgia!
  • Apr, 2024: One paper is accepted by TIST!
  • Mar, 2024: One paper is accepted by NAACL 2024!
  • Jan, 2024: One paper is accepted by WWW 2024!
  • Sep, 2023: One paper is accepted by NIPS 2023!
  • May, 2023: I joined the Tencent AI Lab (@Seattle) as a research intern!
  • Apr, 2023: One paper is accepted by AIED 2023!
  • Jan, 2022: I joined the Data Lab at UGA (DLGA) as a research assistant!

Recent Pre-Prints

  • A Survey on Sparse Autoencoders: Interpreting the Internal Mechanisms of Large Language Models
    Dong Shu, Xuansheng Wu, Haiyan Zhao, Daking Rai, Ziyu Yao, Ninghao Liu, Mengnan Du
    [Paper]

  • Interpreting and Steering LLMs with Mutual Information-based Explanations on Sparse Autoencoders
    Xuansheng Wu, Jiayi Yuan, Wenlin Yao, Xiaoming Zhai, Ninghao Liu
    [Paper]

  • Self-Regularization with Latent Space Explanations for Controllable LLM-based Classification
    Xuansheng Wu, Wenhao Yu, Xiaoming Zhai, Ninghao Liu
    [Paper]

  • Usable XAI: 10 Strategies Towards Exploiting Explainability in the LLM Era
    Xuansheng Wu*, Haiyan Zhao*, Yaochen Zhu*, Yucheng Shi*, Fan Yang, Tianming Liu, Xiaoming Zhai, Wenlin Yao, Jundong Li, Mengnan Du, Ninghao Liu
    [Paper] [Code]

Publications

hpp

Retrieval-enhanced Knowledge Editing in Language Models for Multi-Hop Question Answering
Yucheng Shi, Qiaoyu Tan, Xuansheng Wu, Shaochen Zhong, Kaixiong Zhou, and Ninghao Liu.
CIKM (oral), 2024.
[Paper] [Code]



hpp

From Language Modeling to Instruction Following: Understanding the Behavior Shift in LLMs after Instruction Tuning
Xuansheng Wu, Wenlin Yao, Jianshu Chen, Xiaoman Pan, Xiaoyang Wang, Ninghao Liu, and Dong Yu.
NAACL (oral), 2024.
[Paper] [Code]



hpp

Could Small Language Models Serve as Recommenders? Toward Data-centric Cold-start Recommendation
Xuansheng Wu, Huachi Zhou, Wenlin Yao, Xiao Huang, and Ninghao Liu.
WWW (oral), 2024.
[Paper] [Code]



hpp

Black-box Backdoor Defense via Zero-shot Image Purification
Yucheng Shi, Mengnan Du, Xuansheng Wu, Zihan Guan, Jin Sun, and Ninghao Liu.
NIPS, 2023.
[Paper] [Code]



hpp

Matching Exemplar as Next Sentence Prediction (MeNSP): Zero-shot Prompt Learning for Automatic Scoring in Science Education
Xuansheng Wu, Xinyu He, Tianming Liu, Ninghao Liu, and Xiaoming Zhai.
AIED (oral), 2023.
[Paper] [Code]



hpp

Rethinking the Impacts of Overfitting and Feature Quality on Small-scale Video Classification.
Xuansheng Wu, Feichi Yang, Xinyue Lin, and Tong Zhou.
ACM MM (oral), 2021.
[Paper] [Code]

Projects

DaPy: an easy-to-use data analysis framework for humans.
Xuansheng Wu, and Feichi Yang.
[Website]


Distributed Archery Events Management System.
Xuansheng Wu, Feichi Yang, and Jingjing Guo.
[Website]

Working Experiments

Awards

  • 2024 Travel Award from National Science Foundation
  • 2023 Travel Award from University of Georgia
  • 2021 Tencent Advertising Algorithm Competition - Top 2%
  • 2020 Shanghai Aijian Scholarship - Top 1%
  • 2019 Research Pioneer Award of SUIBE - Top 1%
  • 2019 2nd Prize of China Programming Contest for College Students - Top 5%

Services

  • Journal Reviewer: TNNLS, TKDE, & TOIS.
  • Conference Reviewer: EMNLP 2024, CIKM 2024, FM-Wild Workshop in ICML 2024, & R2-FM Workshop in ICLR 2024.

Teaching

  • 2022 Spring: Teaching Assistant of CSCI4380/6380 Data Mining, University of Georgia.

More about me!

I enjoy sports and traveling with my family and girlfriend. Here are a few of my favorite activities:

  • Archery: I can shoot at least 6 points for 70-meter targets (Olympic Standard).
  • Running: I can run 10KM at pace of 5'40''/KM. I am actively training for my first half-marathon this fall.
  • Basketball: I used to play basketball during high school and undergraduate studies.