Welcome!

Hi! I am a Ph.D. candidate in the department of Electrical and Computer Engineering (ECE) at University of Wisconsin-Madison, advised by Prof. Dimitris Papailiopoulos and Prof. Kangwook Lee. I received my M.S. in 2020 from Seoul National University, where I was advised by Prof. Jungwoo Lee and learned about communication systems. I also received my B.S. in ECE from Seoul National University. I am a recipient of the Korean Government Scholarship Program for Study Overseas.

I am open to ML research, applied scientist, and ML engineering roles in LLMs, post-training, evaluation, and reasoning.

Research Interest

My research centers on improving the generalization and reasoning capabilities of LLMs and foundation models. Key areas include:

Generalization beyond training distributions: compositional, length-based, and easy-to-hard transfer
Algorithmic, mathematical, and abstract reasoning in LLMs
Leveraging self-improvement frameworks, post-training optimization, RLVR, chain-of-thought prompting, and evaluation pipelines to enhance performance, robustness, and adaptability

Selected Publications

Extrapolation by Association: Length Generalization Transfer in Transformers, Neurips 2025 (Spotlight)
Z. Cai, N. Lee, A. Schwarzschild, S. Oymak, D. Papailiopoulos
Self-Improving Transformers Overcome Easy-to-Hard and Length Generalization Challenges, ICML 2025
N. Lee, Z. Cai, A. Schwarzschild, K. Lee, D. Papailiopoulos
Can Mamba Learn How to Learn? A Comparative Study on In-Context Learning Tasks, ICML 2024
J. Park, J. Park, Z. Xiong, N. Lee, J. Cho, S. Oymak, K. Lee, D. Papailiopoulos
Teaching Arithmetic to Small Transformers, ICLR 2024
N. Lee, K. Sreenivasan, J. D. Lee, K. Lee, D. Papailiopoulos
Super Seeds: extreme model compression by trading off storage with computation, ICML 2022 Workshop (oral spotlight)
N. Lee, S. Rajput, J. Sohn, H. Wang, A. Nagle, E. Xing, K. Lee, D. Papailiopoulos

Education

Ph.D. Candidate in Electrical and Computer Engineering, University of Wisconsin–Madison (Sep. 2020 – May 2026, expected)

Advisors: Prof. Dimitris Papailiopoulos & Prof. Kangwook Lee
Research area: LLM post-training, evaluation, reasoning, and generalization

M.S. in Electrical and Computer Engineering, Seoul National University (Mar. 2018 – Aug. 2020)

Advisor: Prof. Jungwoo Lee
Research area: Channel estimation, non-orthogonal multiple access
Thesis: Channel Estimation Using Deep Complex Networks in Massive MIMO-OFDM System

B.S. in Electrical and Computer Engineering, Seoul National University (Mar. 2014 – Feb. 2018)

Summa Cum Laude

Work Experience

Amazon AWS (2025.05 - 2025.08)

Applied Scientist Intern
Mentor: Dr. Dionysis Manousakas
Project: Learning to Improve Without Labels: Diversity as the Key to LLM Self-Improvement
Built self-improving LLM post-training pipelines using model-generated data and analyzed diversity collapse and filtering strategies for performance improvement.

Nanyang Technological University, SNU (2017.06 - 2017.08)

Undergraduate Research Intern
Mentor: Prof. Junsong Yuan
Projects: Hand Gesture Recognition Task Using Deep Learning Based on Google Soli
Improved hand gesture recognition accuracy and implemented a low-latency online recognition system.

Teaching Experience

Graduate Teaching Assistant

(UW-Madison CS/ECE 561) Probability and Information Theory in Machine Learning, 2022 Fall
(UW-Madison CS/ECE/ME 532) Matrix Methods in Machine Learning, 2021 Fall
(SNU) Information Theory, 2019-1
(SNU) Introduction to Communications, 2018-2

Undergraduate Teaching Assistant

(SNU) Foundation of Physics 1, 2015-1, 2016-1,2017-1
(SNU) Foundation of Physics 2, 2015-2, 2016-2, 2017-2

Talks

(02.24.2025) Talk at the FLaNN Seminars
(09.11.2023) Talk at the IFDS Ideas Forum
(08.18.2023) Short Talk at the Simons Institute LLM Workshop - Video Slides