Hanlin Zhang

About

I am a CS PhD candidate at Harvard, advised by Sham Kakade.

I am interested in scientific foundations and social implications of AI.

I received my Master's degree in Machine Learning at CMU, advised by Eric Xing.

Featured Works

Discovering Hierarchical Latent Capabilities of Language Models via Causal Representation Learning
Preprint, 2025. [Paper] [Blog] [Code]
Jikai Jin, Vasilis Syrgkanis, Sham Kakade, Hanlin Zhang

EvoLM: In Search of Lost Language Model Training Dynamics
NeurIPS, 2025. (Oral) [Paper] [Blog] [Code]
Zhenting Qi, Fan Nie, Alexandre Alahi, James Zou, Himabindu Lakkaraju, Yilun Du, Eric Xing, Sham Kakade, Hanlin Zhang

Hanlin Zhang, Depen Morwani, Nikhil Vyas, Jingfeng Wu, Difan Zou, Udaya Ghai, Dean Foster, Sham Kakade

Watermarks in the Sand: Impossibility of Strong Watermarking for Generative Models
ICML, 2024. [Paper] [Website] [Blog] [Code]
Hanlin Zhang, Benjamin L Edelman*, Danilo Francati*, Daniele Venturi, Giuseppe Ateniese, and Boaz Barak

Improved Logical Reasoning of Language Models via Differentiable Symbolic Programming
Findings of ACL, 2023. [Paper] [Code]
Hanlin Zhang*, Jiani Huang*, Ziyang Li, Mayur Naik, Eric Xing

Towards Principled Disentanglement for Domain Generalization
CVPR, 2022. (Oral) [Paper] [Code]
Hanlin Zhang*, Yi-Fan Zhang*, Weiyang Liu, Adrian Weller, Bernhard Schölkopf, Eric Xing

Theme by orderedlist