Stephen Zhang

About Me

Hello! I am a PhD. student in the Department of Mathematics at the University of Toronto under the supervision of Vardan Papyan. Prior to this, I completed my HBSc at University of Toronto St. George Campus as a Mathematics Specialist, Computer Science Major, and Statistics Minor.

I am broadly interested in model compression, specifically neural network pruning, and model interpretability to better develop and understand how pruning performs. My CV can be found here and you can e-mail me at: stephenn.zhang [at] mail.utoronto.ca.

Publications and Preprints

A more detailed description of each can be found under Research.
* denotes equal contribution

  1. OATS: Outlier-Aware Pruning Through Sparse and Low Rank Decomposition
    Stephen Zhang, Vardan Papyan
    International Conference on Learning Representations (ICLR), 2025
    arXiv · GitHub

  2. Sparsest Models Elude Pruning: An Exposé of Pruning’s Current Capabilities
    Stephen Zhang, Vardan Papyan
    International Conference on Machine Learning (ICML), 2024
    arXiv · GitHub

  3. Low-Rank is Required for Pruning LLMs
    Stephen Zhang, Vardan Papyan
    ICLR Workshop on Sparsity in LLMs (SLLM), 2025
    OpenReview

  4. A ‘Catch, Tag, and Release’ Mechanism for Embeddings
    Stephen Zhang*, Mustafa Khan*, Vardan Papyan
    Preprint
    arXiv

Misc.

In my spare time, I enjoy cooking and olympic weightlifting and I am also an avid Formula 1 fan. I was raised in Ottawa and was enrolled in a music program in high school where I played the alto saxophone.

This website was built using a Jekyll theme by Ankit Sultana found here.
The customized cat animation is hand-drawn by Camilla Xue.