Jerry YC Hu's Research Page

Teaching

[Spring 2025,2024] TA, "Advanced Topics on Deep Learning" (CS 496), Northwestern University CS.
[Spring 2026, 2025,2024] TA, "Mathematical Foundations of Machine Learning" (CS 416 & STAT 435), Northwestern University CS & Stats & Data Science.
[Winter 2025, 2022] TA, "Data Science Pipeline" (CS 326), Northwestern University CS.
[Fall 2020] Instructor, "Vibrations, Waves, & Electromagnetism." Notes & recordings upon request, University of Maryland College Park Physics.
[2018-2021] TA for multiple courses (PHYS121, PHYS122, PHYS131, PHYS132, PHYS260, PHYS261, PHYS603...), University of Maryland College Park Physics.
[Spring 2018] TA, "Gravity, S-Matrix & String Theory," NTU Physics.
[Fall 2017] TA, "Quantum Field Theory I," NTU Physics.

Collaboration and Mentoring
(in reverse chronological order)

Mingcheng Lu, Fudan University Math BS'27

Transformer Approximations from ReLUs [arXiv]
A Theoretical Analysis of Discrete Flow Matching Generative Models [arXiv]
Learning Manifold Data with Flow Matching [ICML'26]

Xiwen Zhang, Fudan University Math BS'27

On Structure State-Space Duality [ICML'26]
Learning Constrained Boolean Functions with Softmax Attention [arXiv]

Po-Chiao Lin, NTU Physics BS'25

Universal Approximation, Compositional Generalization, and Algorithm Emulation All In-Context [ICML'26]

Venkat Sripad Ganti, Northwestern Math+Stats+Econ BS'27

Chain-of-Thought Gradient Descent [ICML'26]

Jennifer Yuntong Zhang, University of Toronto BAS'26 Engineering Science -> Yale CS MS (Fall'26)

In‑Context Algorithm Emulation in Fixed‑Weight Transformers [ICLR'26]

Hude Liu, Fudan University Math BS'25

In‑Context Algorithm Emulation in Fixed‑Weight Transformers [ICLR'26]
Attention Mechanism, Max-Affine Partition, and Universal Approximation [NeurIPS'25a]
Universal Approximation with Softmax Attention [ICML'26]

Maojiang Su, University of Science and Technology of China (School of the Gifted Young) BS'25 -> CS PhD at NU (Fall'25)

A Theoretical Analysis of Discrete Flow Matching Generative Models [arXiv]
High-Order Flow Matching:Unified Framework and Sharp Statistical Rates [NeurIPS'25b]
In-Context Deep Learning via Transformer Models [ICML'25b]
Computational Limits of Low-Rank Adaptation (LoRA) for Transformer-Based Models [ICLR'25a]

Zoe Mehta, High School Outreach Student @ Vernon Hills High School -> MIT (Class of 2029)

Fast and Low-Cost Genomic Foundation Models via Outlier Removal [ICML'25a]

Sophia Pi, Northwestern CS+Econ+Math Methods in Social Science BS'26 -> UPenn CS PhD (Fall'26)

Learning Manifold Data with Flow Matching [ICML'26]
On Statistical Rates and Provably Efficient Criteria of Latent Diffusion Transformers (DiTs) [NeurIPS'24b]
On Flow Matching KL Divergence [arXiv]

Chenghao Qiu, Tianjin University CS BS'25 -> CS PhD study at TAMU (Fall'25)

Fast and Low-Cost Genomic Foundation Models via Outlier Removal [ICML'25a]

Thomas Yuan-Lung Lin, High School Outreach Student @ WLSH'23 -> NTU (transferred) -> U. of Washington Physics (Class of 2027)

- Latent Variable Estimation in Bayesian Black-Litterman Models [ICML'25e]
- On Computational Limits of Modern Hopfield Models: A Fine-Grained Complexity Analysis [ICML'24a]

Stephen Cheng, Northwestern EE BS'25 + CS MS'25 -> CS PhD at UMD (Fall'25)

Financial Data Prediction Models, Statistical Theory of Diffusion Models and Transformer

Teng-Yun Hsiao, NTU Physics BS'26

In-Context Learning as Conditioned Associative Memory Retrieval [ICML'25c]
Uniform Memory Retrieval with Larger Capacity for Modern Hopfield Models [ICML'24d]

Wei-Po Wang, NTU Physics BS'24 -> Phys PhD at Johns Hopkins University (Fall'25)

Fundamental Limits of Prompt Tuning Transformers: Universality, Capacity and Efficiency [ICLR'25b]
Outlier-Efficient Hopfield Layers for Large Transformer-Based Models [ICML'24b]

Morris Huang, NTU Physics MS'24 -> CS PhD at North Carolina, Chapel Hill (Fall'25)

On Statistical Rates of Conditional Diffusion Transformers: Approximation, Estimation and Minimax Optimality [ICLR'25c]
BiSHop: Bi-Directional Cellular Learning for Tabular Data with Generalized Sparse Hopfield Model [ICML'24]

Hong-Yu Chen, NTU Physics MS'24 -> CS PhD study at NU (Fall'24)

- Universal Approximation of Softmax Attention [ICML'26]
- Outlier-Efficient Hopfield Layers for Large Transformer-Based Models [ICML'24b]

Yi-Chen Lee, NTU Physics BS'26 -> CS PhD study at Johns Hopkins University (Fall'26)

High-Order Flow Matching:Unified Framework and Sharp Statistical Rates [NeurIPS'25b]
On Statistical Rates of Conditional Diffusion Transformers: Approximation, Estimation and Minimax Optimality [ICLR'25c]

Bo-Yu Chen, High School Outreach Student @ HSNU'23 -> NTU Physics + CS (Class of 2027) with NTU Fu Bell Scholarship (Highest Distinction across University)

- Nonparametric Modern Hopfield Models [ICML'25c]
- STanHop: Sparse Tandem Hopfield Model for Memory-Enhanced Time Series Prediction [ICLR'24]
- On Sparse Modern Hopfield Model [NeurIPS'23]

Chenwei Xu, MSCS'24 at NU -> Stats & DS PhD study at NU (Fall'24)

- BiSHop: Bi-Directional Cellular Learning for Tabular Data with Generalized Sparse Hopfield Model [ICML'24c]
- On Sparse Modern Hopfield Model [NeurIPS'23]
- Beyond PID Controllers: PPO with Neuralized PID Policy for Proton Beam Intensity Control in Mu2e [ML4Phys Workshop @ NeurIPS'23] [READS Collaboration, Fermilab]
- Feature Programming for Multivariate Time Series Prediction [ICML'23]

Dennis Wu, MSCS'24 at NU -> CS PhD study at NU (Fall'24)

- Provably Optimal Memory Capacity for Modern Hopfield Models [NeurIPS'24]
- Uniform Memory Retrieval with Larger Capacity for Modern Hopfield Models [ICML'24d]
- STanHop: Sparse Tandem Hopfield Model for Memory-Enhanced Time Series Prediction [ICLR'24]
- On Sparse Modern Hopfield Model [NeurIPS'23]

Zhenyu Pan, MSECE'24 at U. of Rochester -> CS PhD study at NU (Fall'24)

Zhenji Wang, UMD'21 Math -> MS'23 at Columbia U. -> ML PhD study at U. of Tsukuba

- Differential Geometry and Heat Kernel Expansion, 2021 Spring

A nice PhD checklist from Aaditya Ramdas

What are expected for graduate research

A Note to Prospective Students by Bert Huang
A Note to Prospective Students by Sanmi Koyejo

An open letter to graduate students and other procrastinators: it’s time to write (from Dennis J. Hazelett)

https://www.nature.com/articles/s41587-025-02584-1

10 easy ways to fail a Ph.D. by from Matt Might

https://matt.might.net/articles/ways-to-fail-a-phd

“To boldly go where no man has gone before.”

By far, Star Trek has one of the most famous opening narrations in the history of the filming industry. Resonating a feeling of the 1960’s futuristic optimism, the iconic phrase was first spoken by William Shatner who starred as Captain James T. Kirk in the original series.

“Space: the final frontier. These are the voyages of the starship Enterprise. Its five-year mission: to explore a strange new world, to seek out new life and new civilizations, to boldly go where no man has gone before.”

Publications

Please see for [Google Scholar] the latest publications.

2026

On Structure State-Space Duality
Jerry Yao-Chieh Hu*, Xiwen Zhang*, Ali ElSheikh, Weimin Wu, Han Liu
International Conference on Machine Learning 2026. ICML'26

Univeral Approximation, Compositional Generalization, and Algorithm Emulation All In-Context
Hong-Yu Chen*, Po-Chiao Lin*, Jerry Yao-Chieh Hu*, Maojiang Su, Han Liu
International Conference on Machine Learning 2026. ICML'26

Chain-of-Thought Gradient Descent
Hong-Yu Chen*, Venkat Sripad Ganti*, Jerry Yao-Chieh Hu*, Hude Liu, Han Liu
International Conference on Machine Learning 2026. ICML'26

Universal Approximation with Softmax Attention
Jerry Yao-Chieh Hu*, Hude Liu*, Hong-Yu Chen*, Weimin Wu, Han Liu
International Conference on Machine Learning 2026. ICML'26

Genome-Factory: An Integrated Python Library for Tuning, Deploying and Interpreting Genomic Models
Weimin Wu*, Xuefeng Song*, Yibo Wen*, Qinjie Lin, Zhihan Zhou, Jerry Yao-Chieh Hu, Zhong Wang, Han Liu
International Conference on Machine Learning 2026. ICML'26

Learning Manifold Data with Flow Matching
Sophia Pi*, Mingcheng Lu*, Maojiang Su, Weimin Wu, Jerry Yao-Chieh Hu, Han Liu
International Conference on Machine Learning 2026. ICML'26

In-Context Algorithm Emulation in Fixed-Weight Transformers
Jerry Yao-Chieh Hu*, Hude Liu*, Jennifer Zhang*, Han Liu
International Conference on Learning Representations 2026. ICLR'26

2025

Attention Mechanism, Max-Affine Partition, and Universal Approximation
Hude Liu*, Jerry Yao-Chieh Hu*, Zhao Song, Han Liu
Advances in Neural Information Processing Systems 2025. NeurIPS'25

High-Order Flow Matching: Unified Framework and Sharp Statistical Rates
Maojiang Su*, Jerry Yao-Chieh Hu*, Yi-Chen Lee*, Ning Zhu, Jui-Hui Chung, Shang Wu, Zhao Song, Minshuo Chen, Han Liu
Advances in Neural Information Processing Systems 2025. NeurIPS'25

Pareto-Optimal Energy Alignment for Designing Nature-Like Antibodies
Yibo Wen, Chenwei Xu, Jerry Yao-Chieh Hu, Kaize Ding, Han Liu
Advances in Neural Information Processing Systems 2025. NeurIPS'25

Mind the Inconspicuous: Revealing the Hidden Weakness in Aligned LLMs' Ethical Boundaries
Jiahao Yu*, Haozheng Luo*, Jerry Yao-Chieh Hu, Yan Chen, Wenbo Guo, Han Liu, Xinyu Xing
USENIX Security Symposium 2025. USENIX Security'25

Fast and Low-Cost Genomic Foundation Models via Outlier Removal
Robin Luo*, Chenhao Qiu*, Zhihan Zhuo, Maojiang Su, Zoe Mehta, Jerry Yao-Chieh Hu, Han Liu
International Conference on Machine Learning 2025. ICML'25

Latent Variable Estimation in Bayesian Black-Litterman Models
Thomas Lin, Jerry Yao-Chieh Hu, Paul W. Chiou, Peter Lin
International Conference on Machine Learning 2025. ICML'25

In-Context Deep Learning via Transformer Models
Weimin Wu*, Maojiang Su*, Jerry Yao-Chieh Hu*, Zhao Song, Han Liu
International Conference on Machine Learning 2025. ICML'25

In-Context Learning as Conditioned Associative Memory Retrieval
Weimin Wu*, Teng-Yun Hsiao*, Jerry Yao-Chieh Hu*, Wenxin Zhang, Han Liu
International Conference on Machine Learning 2025. ICML'25

Nonparametric Modern Hopfield Models
Jerry Yao-Chieh Hu, Bo-Yu Chen, Dennis Wu, Feng Ruan, Han Liu
International Conference on Machine Learning 2025. ICML'25

On Statistical Rates of Conditional Diffusion Transformers
Jerry Yao-Chieh Hu*, Weimin Wu*, Yu-Chao Huang*, Yi-Chen Lee*, Minshuo Chen, Han Liu
International Conference on Learning Representations 2025. ICLR'25

Computational Limits of Low-Rank Adaptation (LoRA) for Transformer-Based Models
Jerry Yao-Chieh Hu, Maojiang Su, En-Jui Kuo, Zhao Song, Han Liu
International Conference on Learning Representations 2025. ICLR'25

Fundamental Limits of Prompt Tuning: Universality, Capacity and Efficiency
Jerry Yao-Chieh Hu*, Wei-Po Wang*, Chenyang Li, Ammar Gilani, Zhao Song, Han Liu
International Conference on Learning Representations 2025. ICLR'25

2024

On Statistical Rates and Provably Efficient Criteria of Latent Diffusion Transformers (DiTs)
Jerry Yao-Chieh Hu*, Weimin Wu*, Zhao Song, Han Liu
Advances in Neural Information Processing Systems 2024. NeurIPS'24

Provably Optimal Memory Capacity for Modern Hopfield Models
Jerry Yao-Chieh Hu*, Dennis Wu*, Han Liu
Advances in Neural Information Processing Systems 2024. NeurIPS'24

Outlier-Efficient Hopfield Layers for Large Transformer-Based Models
Jerry Yao-Chieh Hu*, Pei-Hsuan Chang*, Robin Luo*, Hong-Yu Chen, Weijian Li, Wei-Po Wang, Han Liu
International Conference on Machine Learning 2024. ICML'24

BiSHop: Bi-Directional Cellular Learning for Tabular Data with Generalized Sparse Modern Hopfield Model
Chenwei Xu*, Yu-Chao Huang*, Jerry Yao-Chieh Hu*, Weijian Li, Ammar Gilani, Hsu-Sheng Goan, Han Liu
International Conference on Machine Learning 2024. ICML'24

Uniform Memory Retrieval with Larger Capacity for Modern Hopfield Models
Dennis Wu*, Jerry Yao-Chieh Hu*, Teng-Yun Hsiao, Han Liu
International Conference on Machine Learning 2024. ICML'24

On Computational Limits of Modern Hopfield Models: A Fine-Grained Complexity Analysis
Jerry Yao-Chieh Hu*, Thomas Lin*, Zhao Song, Han Liu
International Conference on Machine Learning 2024. ICML'24

STanHop: Sparse Tandem Hopfield Model for Memory-Enhanced Time Series Prediction
Dennis Wu*, Jerry Yao-Chieh Hu*, Weijian Li*, Bo-Yu Chen, Han Liu
International Conference on Learning Representations 2024. ICLR'24

2023

On Sparse Modern Hopfield Model
Jerry Yao-Chieh Hu, Donglin Yang, Dennis Wu, Chenwei Xu, Bo-Yu Chen, Han Liu
Advances in Neural Information Processing Systems 2023. NeurIPS'23

Feature Programming for Multivariate Time Series Prediction
Alex Reneau*, Jerry Yao-Chieh Hu*, Chenwei Xu, Weijian Li, Ammar Gilani, Han Liu
International Conference on Machine Learning 2023. ICML'23

Ising-traffic: Using Ising Machine Learning to Predict Traffic Congestion under Uncertainty
Zhenyu Pan, Anshujit Sharma, Jerry Yao-Chieh Hu, Zhuo Liu, Ang Li, Han Liu, Michael Huang, Tony Geng
Proceedings of the AAAI Conference on Artificial Intelligence 2023. AAAI'23

FPGA Architectures for Distributed ML Systems for Real-time Beam Loss De-blending
M. A. Ibrahim et al., including Jerry Yao-Chieh Hu (Fermilab READS Collaboration)
International Beam Instrumentation Conference, Report number: FERMILAB-CONF-23-281-AD. IBIC'23

Disentangling Beam Losses in the Fermilab Main Injector Enclosure Using Real-Time Edge AI
K. J. Hazelwood et al., including Jerry Yao-Chieh Hu (Fermilab READS Collaboration)
International Conference on Accelerator and Large Experimental Physics Control Systems 2023.

Google Sites

Report abuse