About Me

I am a Senior Researcher at Microsoft Research AI Frontiers, working with Prof. Dimitris Papailiopoulos. I received my PhD degree in Computer Science from the University of Wisconsin-Madison in 2025, where I was advised by Prof. Kangwook Lee. I received my Master's degree in Statistics from UW-Madison in 2020, where I was advised by Prof. Miaoyan Wang. Before joining UW-Madison, I completed my Bachelor's degree in Statistics from the School of Mathematical Sciences at Nankai University in 2019.

Selected Publications

(* equal contribution, † equal advising)

For a full list of publications, please see my Google Scholar.

LLM

You Don't Need to Run Every Eval
Yuchen Zeng, Dimitris Papailiopoulos
Under Review
[Paper] [Code] [Dataset] [Website]

ReasoningLLM

Memento: Teaching LLMs to Manage Their Own Context
Vasilis Kontonis, Yuchen Zeng, Shivam Garg, Lingjiao Chen, Hao Tang, Ziyan Wang, Ahmed Awadallah, Eric Horvitz, John Langford, Dimitris Papailiopoulos
2026 Conference on Language Modeling (COLM)
[Paper] [Code] [Dataset] [Blog]

LLMICLReasoningTheory

Demonstrations, CoT, and Prompting: A Theoretical Analysis of ICL
Xuhan Tong, Yuchen Zeng†, Jiawei Zhang†
Under Review
[Paper] [Code]

LLMReasoning

ReJump: A Tree-Jump Representation for Analyzing and Improving LLM Reasoning
Yuchen Zeng*, Shuibai Zhang*, Wonjun Kang*, Shutong Wu, Lynnix Zou, Ying Fan, Heeju Kim, Ziqian Lin, Jungtaek Kim, Hyung Il Koo, Dimitris Papailiopoulos, Kangwook Lee
International Conference on Machine Learning 2026 (ICML)
[Paper] [Code] [Summary]

LLMDiffusion LLMBenchmark

ParallelBench: Understanding the Trade-offs of Parallel Decoding in Diffusion LLMs
Wonjun Kang*, Kevin Galim*, Seunghyuk Oh*, Minjae Lee, Yuchen Zeng, Shuibai Zhang, Coleman Hooper, Yuezhou Hu, Hyung Il Koo, Nam Ik Cho, Kangwook Lee
2026 International Conference on Learning Representation (ICLR)
[Paper] [Code] [Summary] [Website]

PEFTState Space ModelLLM

State-offset Tuning: State-based Parameter-Efficient Fine-Tuning for State Space Models
Wonjun Kang*, Kevin Galim*, Yuchen Zeng*, Minjae Lee, Hyung Il Koo, Nam Ik Cho
2025 Annual Conference of the Association for Computational Linguistics (ACL)
[Paper] [Code]

LLMTabular LearningTransformerICL

TabFlex: Scaling Tabular Learning to Millions with Linear Attention
Yuchen Zeng*, Tuan Dinh*, Wonjun Kang, Andreas Mueller
International Conference on Machine Learning 2025 (ICML) Spotlight (Top 2.6%)
[Paper] [Code]

PEFTState Space ModelLLM

Parameter-Efficient Fine-Tuning of State Space Models
Kevin Galim*, Wonjun Kang*, Yuchen Zeng*, Hyung Il Koo, Kangwook Lee
International Conference on Machine Learning 2025 (ICML)
[Paper] [Code] [Summary]

LLMMultimodalICLBenchmark

Can MLLMs Perform Text-to-Image In-Context Learning?
Yuchen Zeng*, Wonjun Kang*, Yicong Chen, Hyung Il Koo, Kangwook Lee
2024 Conference on Language Modeling (COLM)
[Paper] [Code] [Summary] [45-Minute Talk]

PEFTLLMTheory

The Expressive Power of Low-Rank Adaptation
Yuchen Zeng, Kangwook Lee
2024 International Conference on Learning Representation (ICLR)
[Paper] [Code] [Summary] [45-Minute Talk in Chinese]

FairnessFederated LearningTheory

Federated Learning with Local Fairness Constraints
Yuchen Zeng, Hongxu Chen, Kangwook Lee
2023 IEEE International Symposium on Information Theory (ISIT)
[Paper]

Fairness

Equal Improvability: A New Fairness Notion Considering the Long-Term Impact
Ozgur Guldogan*, Yuchen Zeng*, Jy-yong Sohn, Ramtin Pedarsani, Kangwook Lee
2023 International Conference on Learning Representation (ICLR)
[Paper] [Code] [Summary] [5-Minute Video] [Article]

LLMPEFTMultimodalTabular Learning

LIFT: Language-Interfaced FineTuning for Non-Language Machine Learning Tasks
Tuan Dinh*, Yuchen Zeng*, Ruisu Zhang, Ziqian Lin, Michael Gira, Shashank Rajput, Jy-yong Sohn, Dimitris Papailiopoulos, Kangwook Lee
Neural Information Processing Systems 2022 (NeurIPS)
[Paper] [Code] [Summary] [5-Minute Video] [Website] [45-Minute Talk in Chinese]

Tensor LearningTheory

Multiway clustering via tensor block models
Miaoyan Wang, Yuchen Zeng
Neural Information Processing Systems 2019 (NeurIPS)
[Paper] [Code] [3-Minute Video]

Services and Leadership

Recipient of 2024 COLM DEI Travel Scholarship
Selected to participate in Future Leaders Summit
Organizer of MLOPT Idea Seminar
Conference reviewer: NeurIPS, ICML, ICLR, TMLR, AISTATS

Work Experience

Meta, Fall 2024
Mentored by Dr. Yipin Zhou and Dr. Yixin Nie

Microsoft, Summer 2024
Mentored by Dr. Andreas Christian Müller

MIT-IBM Watson AI Lab, Summer 2022
Mentored by Dr. Mikhail Yurochkin and Prof. Justin Solomon