Can Yaras

PhD student, University of Michigan - Ann Arbor

Joined

May 2022

Names

Can Yaras (Preferred)

Emails

****@umich.edu (Confirmed)

Personal Links

Homepage

Google Scholar

DBLP

Career & Education History

PhD student

University of Michigan - Ann Arbor (umich.edu)

2021 – 2026

Undergrad student

Duke University (duke.edu)

2017 – 2021

Advisors, Relations & Conflicts

PhD Advisor

Qing Qu

2021 – 2026

PhD Advisor

Laura Balzano

2021 – 2026

Expertise

deep learning

nonconvex optimization

2021 – 2023

Publications

Towards Understanding Out-of-Distribution Generalization for In-Context Learning via Low-Dimensional Subspaces
Soo Min Kwon, Alec S. Xu, Can Yaras, Laura Balzano, Qing Qu
- CauScale 2026
- Readers: Everyone
Out-of-Distribution Generalization of In-Context Learning: A Low-Dimensional Subspace Perspective
Soo Min Kwon, Alec S. Xu, Can Yaras, Laura Balzano, Qing Qu
- AISTATS 2026 Poster
- Readers: Everyone
Linearly Separable Features in Shallow Nonlinear Networks: Width Scales Polynomially with Intrinsic Data Dimension
Alec S. Xu, Can Yaras, Peng Wang, Qing Qu
- AISTATS 2026 Poster
- Readers: Everyone
Zero-Shot Conversion to Monarch-Structured Attention
Can Yaras, Alec S. Xu, Pierre Abillama, Changwoo Lee, Laura Balzano
- ES-FoMo III
- Readers: Everyone
MonarchAttention: Zero-Shot Conversion to Fast, Hardware-Aware Structured Attention
Can Yaras, Alec S Xu, Pierre Abillama, Changwoo Lee, Laura Balzano
- NeurIPS 2025 spotlight
- Readers: Everyone
An Overview of Low-Rank Structures in the Training and Adaptation of Large Models
Laura Balzano, Tianjiao Ding, Benjamin D. Haeffele, Soo Min Kwon, Qing Qu, Peng Wang, Zhangyang Wang, Can Yaras
- CoRR 2025
- Readers: Everyone
Understanding How Nonlinear Networks Create Linearly Separable Features for Low-Dimensional Data
Alec S Xu, Can Yaras, Peng Wang, Qing Qu
- CPAL 2025 (Recent Spotlight Track)
- Readers: Everyone
Understanding How Nonlinear Layers Create Linearly Separable Features for Low-Dimensional Data
Alec S. Xu, Can Yaras, Peng Wang, Qing Qu
- CoRR 2025
- Readers: Everyone
Explaining and Mitigating the Modality Gap in Contrastive Multimodal Learning
Can Yaras, Siyi Chen, Peng Wang, Qing Qu
- CPAL 2025 (Proceedings Track) Poster
- Readers: Everyone
Compressible Dynamics in Deep Overparameterized Low-Rank Learning & Adaptation
Can Yaras, Peng Wang, Laura Balzano, Qing Qu
- ICML 2024 Oral
- Readers: Everyone

View all 21 publications

Can Yaras

Names

Emails

Personal Links

Career & Education History

Advisors, Relations & Conflicts

Expertise

Publications

Co-Authors