Shiyun Xu

Joined

September 2020

Names

Shiyun Xu (Preferred)

Emails

****@sas.upenn.edu (Confirmed)

****@gmail.com (Confirmed)

Personal Links

Homepage

DBLP

Career & Education History

PhD student

University of Pennsylvania, University of Pennsylvania (upenn.edu)

2021 – 2025

MS student

University of Pennsylvania (upenn.edu)

2019 – 2021

Advisors, Relations & Conflicts

No relations added

Expertise

No areas of expertise listed

Publications

FlowNIB: An Information Bottleneck Analysis of Bidirectional vs. Unidirectional Language Models
Md Kowsher, Nusrat Jahan Prottasha, Shiyun Xu, Shetu Mohanto, Niloofar Yousefi, Ozlem Garibay, Chen Chen
- ICLR 2026 Poster
- Readers: Everyone
Convex Dominance in Deep Learning I: A Scaling Law of Loss and Learning Rate
Zhiqi Bu, Shiyun Xu, Jialin Mao
- ICLR 2026 Poster
- Readers: Everyone
Adaptive parameter-efficient fine-tuning via multi-task optimization on subset selection
Shiyun Xu, Zhiqi Bu
- ICLR 2026 Conference Withdrawn Submission
- Readers: Everyone
Not all parameters are equal: a Hessian informed differential learning rate for deep learning
Shiyun Xu, Zhiqi Bu, Yiliang Zhang, Ian Barnett
- ICLR 2025 Conference Withdrawn Submission
- Readers: Everyone
Gradient descent with generalized Newton’s method
Zhiqi Bu, Shiyun Xu
- ICLR 2025 Poster
- Readers: Everyone
Accelerated Deep Active Learning with Graph-based Sub- Sampling
Dan Kushnir, Shiyun Xu
- Accepted by TMLR
- Readers: Everyone
Scalable and Efficient Training of Large Convolutional Neural Networks with Differential Privacy
Zhiqi Bu, Jialin Mao, Shiyun Xu
- Published: 31 Oct 2022, Last Modified: 08 Feb 2026
- NeurIPS 2022 Accept
- Readers: Everyone
Sparse Neural Additive Model: Interpretable Deep Learning with Feature Selection via Group Sparsity
Shiyun Xu, Zhiqi Bu, Pratik Chaudhari, Ian J. Barnett
- Published: 25 Mar 2022, Last Modified: 14 Jan 2026
- ICLR 2022 PAIR^2Struct Poster
- Readers: Everyone
Sparse Neural Additive Model: Interpretable Deep Learning with Feature Selection via Group Sparsity
Shiyun Xu, Zhiqi Bu, Pratik Chaudhari, Ian J. Barnett
- 2022 (modified: 29 Jan 2023)
- CoRR 2022
- Readers: Everyone
Scalable and Efficient Training of Large Convolutional Neural Networks with Differential Privacy
Zhiqi Bu, Jialin Mao, Shiyun Xu
- 2022 (modified: 16 Apr 2023)
- CoRR 2022
- Readers: Everyone

View all 13 publications