Hello Friend!
I am a fourth year Ph.D. student in the Decision, Risk, and Operations department at Columbia University. I am fortunate to be advised by Prof. Hongseok Namkoong and Prof. Tianyi Peng.
I am broadly interested in training and deploying agentic systems. My research focuses on developing AI agents for real-world sequential decision-making under uncertainty.
Prior to Columbia, I received my B.A. in Mathematics from Princeton University.
Preprints
Speculative Actions: A Lossless Framework for Faster Agentic Systems.
Naimeng Ye*, Arnav Ahuja*, Georgios Liargkovas*, Yunan Lu*, Kostis Kaffes, and Tianyi Peng.
arXiv:2510.04371 [cs.AI].
Under review
Exchangeable Sequence Models Can Naturally Quantify Uncertainty Over Latent Concepts.
Naimeng Ye and Hongseok Namkoong.
arXiv:2408.03307 [stat.ML].
Submitted to Opereations Research
Differences-in-Neighbors for Network Interference in Experiments.
Tianyi Peng, Naimeng Ye, and Andrew Zheng. (α−β order)
arXiv:2503.02271 [cs.SI].
ACM EC 2025
PersonalLLM: Tailoring LLMs to Individual Preferences.
Tom Zollo, Andrew Siah, Naimeng Ye, Ang Li, and Hongseok Namkoong.
arXiv:2409.20296 [cs.LG].
ICLR 2025
AI Agents for Web Testing: A Case Study in the Wild
Naimeng Ye, Xiao Yu, Ruize Xu, Tianyi Peng, and Zhou Yu
arXiv:2509.05197 [cs.SE].
Working Projects
Algorithmic Learning for Sequence Models
with Priyank Agrawal and Hongseok Namkoong
SynthTools: A Framework for Scaling Synthetic Tools for Agent Development
with Tommaso Castellani, Daksh Mittal, Thomson Yen, and Hongseok Namkoong
Service
Reviewer for ICLR 2026, ICLR 2025, AISTATS 2026, and NeurIPS 2024.