|
Mingda Wan About MeHi! I am Mingda Wan(万明达), a fourth-year undergraduate student at Anhui University. Previously, I was fortunate to be advised by Zhao Song at UC Berkeley, Yingyu Liang and Zhenmei Shi at UW–Madison, for machine learning theory and generative models. Currently, I am looking for further research opportunities and an MPhil or PhD position in 2025 Fall. If you believe I align with your research interests, please feel free to reach out via email. Email / WeChat / CV / Google Scholar / Github |
|
DreamI aspire to provide insightful theoretical analysis of machine learning methods and to leverage those theoretical understanding to guide the design and optimization of future techniques. |
Publications* denote alphabetical order |
|
NRFlow: Towards Noise-Robust Generative Modeling via High-Order Flow Matching
Bo Chen*, Chengyue Gong*, Xiaoyu Li*, Yingyu Liang*, Zhizhou Sha*, Zhenmei Shi*, Zhao Song*, Mingda Wan*, Xugang Ye* UAI 2025 Camera Ready Paper Preview |
|
Attention Scheme Inspired Softmax Regression
Zhihang Li*, Zhizhou Sha*, Zhao Song* and Mingda Wan* ICLR 2025 Workshop Camera Ready Paper |
|
An Improved Sample Complexity for Rank-1 Matrix Sensing
Zhihang Li*, Zhizhou Sha*, Zhao Song* and Mingda Wan* ICLR 2025 Workshop Camera Ready Paper |
|
High-Order Matching for One-Step Shortcut Diffusion Models
Bo Chen*, Chengyue Gong*, Xiaoyu Li*, Yingyu Liang*, Zhizhou Sha*, Zhenmei Shi*, Zhao Song* and Mingda Wan* ICLR 2025 Workshop Camera Ready Paper |
|
Theoretical Constraints on the Expressive Power of RoPE-based Tensor Attention Transformers
Xiaoyu Li*, Yingyu Liang*, Zhenmei Shi*, Zhao Song* and Mingda Wan* arXiv preprint arXiv:2412.18040 arXiv Link |
|
HOFAR: High-Order Augmentation of Flow Autoregressive Transformers
Yingyu Liang*, Zhizhou Sha*, Zhenmei Shi*, Zhao Song* and Mingda Wan* arXiv preprint arXiv:2503.08032 arXiv Link |
|
Force Matching with Relativistic Constraints: A Physics-Inspired Approach to Stable and Efficient Generative Modeling
Yang Cao*, Bo Chen*, Xiaoyu Li*, Yingyu Liang*, Zhizhou Sha*, Zhenmei Shi*, Zhao Song* and Mingda Wan* arXiv preprint arXiv:2502.08150 arXiv Link |
Academic Service |
|
External Reviewer
Research Grants Council (Hong Kong) |
|
Reviewer
ICCV 2025 KDD 2024 |
|
Design and source code from Jon Barron's website. |