Hi, I am Zihang. I am a researcher at UC Berkeley and International Computer Science Institute, advised by Prof. Michael Mahoney. I also have the privilege to work closely with Prof. Yaoqing Yang from Dartmouth College and Shiwei Liu from MPI-IS. I previously obtained my Master’s degree in EECS at UC Berkeley.

My research focus is to understand and improve the transparency and efficiency of learning models. I am particularly interested in:

Understanding the mechanisms and behaviors of AI systems, such as generalization, uncertainty, and reasoning of LLMs, and using principled approaches to improve the transparency and efficiency.
Leveraging the geometry of deep learning such as low-rank subspaces and loss landscapes, to develop new (numerical) algorithms and frameworks for training large-scale AI systems and solving numerical problems.

My research draws inspirations from high-dimensional statistics, random matrix theory and (randomized) numerical linear algebra.

🔥 News

2026.02: Excited to share our recent works bridging spectral analysis and ML: AutoSpec – a neural network framework to discover iterative spectral algorithms for NLA and optimization; HTMuon – improving Muon via heavy-tailed spectral correction.
2025.06: Started my research engineer position at ICSI to work on numerical algorithms and deep learning.
2025.05: Our paper “Principal Weights Emerge after Rank Reduction for Reasoning-Focused Supervised Fine-Tuning” has been accepted to ICML 2025.
2024.11: Gave a presentation at EMNLP 2024 on foundation model diagnosis, check out the live recording here.
2024.09: Excited to share that our work “Model Balancing Helps Low-data Training and Fine-tuning” is accepted by EMNLP 2024 as Oral Presentation.