I am a researcher in the machine learning department at NEC Labs North America. I completed my PhD in mathematics at Stanford University in 2023, co-advised by James Zou and Lexing Ying. In the past, I worked broadly on developing machine learning algorithms with theoretical guarantees. My current interest is understanding how LLMs work using both theoretical and empirical tools.

I can be reached at zach at nec-labs dot com.

Selected Publications

Subgroup Discovery with the Cox Model
Zachary Izzo, Iain Melvin
ICML 2026

Quantitative Bounds for Length Generalization in Transformers
Zachary Izzo*, Eshaan Nichani*, Jason D. Lee
ICLR 2026 (Oral, top 1% of submissions)

Solving Inverse Problems via Diffusion-Based Priors: An Approximation-Free Ensemble Sampling Approach
Haoxuan Chen, Yinuo Ren, Martin Renqiang Min, Lexing Ying, Zachary Izzo
TMLR 2026

Monitoring AI-Modified Content at Scale: A Case Study on the Impact of ChatGPT on AI Conference Peer Reviews [Article 1] [2] [3] [4]
Weixin Liang*, Zachary Izzo*, Yaohui Zhang*, et al.
ICML 2024 (Oral, top 1% of submissions)

Provable Membership Inference Privacy
Zachary Izzo, Jinsung Yoon, Sercan Arik, James Zou
TMLR 2024

Continuous-in-Time Limit for Bayesian Bandits
Yuhua Zhu, Zachary Izzo, Lexing Ying
JMLR 2023

Data-Driven Subgroup Identification for Linear Regression
Zachary Izzo, Ruishan Liu, James Zou
ICML 2023

How to Learn when Data Gradually Reacts to Your Model
Zachary Izzo, James Zou, Lexing Ying
AISTATS 2022

Dimensionality Reduction for Wasserstein Barycenter
Zachary Izzo, Sandeep Silwal, Samson Zhou
NeurIPS 2021

How to Learn when Data Reacts to Your Model: Performative Gradient Descent
Zachary Izzo, Lexing Ying, James Zou
ICML 2021

Borrowing From the Future: Addressing Double Sampling in Model-free Control
Yuhua Zhu, Zachary Izzo, Lexing Ying
MSML 2021

Approximate Data Deletion from Machine Learning Models [Article]
Zachary Izzo, Mary Anne Smart, Kamalika Chaudhuri, James Zou
AISTATS 2021

Preprints/Workshop Papers

Tracing LLM Behavior to the Training Data with Empirical Next-Token Distributions
Zachary Izzo
In submission 2026

Causal Survival Forests with Negative Controls
Zijun Gao, Kyoungeui Hong, Leyi Ma, Qianli Wu, Zachary Izzo, Ruishan Liu
In submission 2026

To Err Is Human: Systematic Quantification of Errors in Published AI Papers via LLM Analysis
Federico Bianchi*, Yongchan Kwon*, Zachary Izzo*, Linjun Zhang, James Zou
arXiv 2026

A Theoretical Study of Dataset Distillation
Zachary Izzo, James Zou
NeurIPS M3L 2023

Importance Tempering: Group Robustness for Overparameterized Models
Yiping Lu, Wenlong Ji, Zachary Izzo, Lexing Ying
arXiv 2022