Zimo Wang

Hi, I am Zimo Wang 王子墨, a Data Science Ph.D. candidate at University of California, San Diego. I’m very lucky to be advised by Professor Tzu-Mao Li, who is both supportive and has a great sense of humor. Together, we explore problems in computer graphics, vision, and beyond, inspired by statistical methods and elegant mathematical ideas.

I am excited about conducting research that combines useful math with real-world applications. We are also open to exploring new directions and connecting dots that others overlooked. Feel free to reach out if you’re interested in my research or would like to chat!

Experience

2023 - Now: Data Science PhD Candidate, University of California San Diego

2026 Summer: Research Intern at ByteDance, San Jose, advised by Gordon Guocheng Qian and Chongyang Ma. Worked on distillation for video generation.

2025 Summer: Research Intern at Adobe Research, advised by Yuting Yang and Jiawen Chen. Worked on generating color mappings from image pairs.

2023 - 2026: Data Science M.S. (Master's along the way), University of California San Diego

2019 - 2023: Majoring in Geographical Information Science, Statistics, Zhejiang University

2019 - 2023: Minoring in Computer Science and Technology, Zhejiang University

Selected Research

PDMD: Projected Distribution Matching Distillation for Video Diffusion Models

Project Page | Paper | Code

Figure 3: PDMD projects the DMD update perpendicular to the student-critic endpoint residual

DMD update target with visible artifacts — DMD target

PDMD update target with fewer artifacts — PDMD target

Zimo Wang, Junkun Yuan, Angtian Wang, Haotian Yang, Canyu Zhang, Siyuan Yuan, Xingchang Huang, Bo Liu, Yizhi Wang, Yiding Yang, Chongyang Ma, Gordon Guocheng Qian.
arXiv preprint. 2026.

tl;dr: PDMD improves the diffusion distillation with a one-line projection that filters student's training error. Nothing else changes: no auxiliary loss, no additional network, no extra model pass.

Distance Marching for Generative Modeling

Project Page | Paper | Code | Advisor: Tzu-Mao Li

Distance Marching follows a distance field toward the data manifold, compared with flow matching

Zimo Wang, Ishit Mehta, Haolin Lu, Xunpeng Huang, Chung-En Sun, Ge Yan, Tsui-Wei Weng, Tzu-Mao Li.
Conference on Neural Information Processing Systems. 2026.

tl;dr: Distance-field modeling enables image generation without time conditioning, with losses that focus on closer targets and two principled samplers that guide noise toward the data manifold faster.

HotSpot: Signed Distance Function Optimization with an Asymptotically Sufficient Condition

Project Page | Paper | Code | Advisor: Tzu-Mao Li

Zimo Wang, Cheng Wang, Taiki Yoshino, Sirui Tao, Ziyang Fu, Tzu-Mao Li. * indicates equal contribution.
Conference on Computer Vision and Pattern Recognition (Highlight). 2025.

tl;dr: Conventional SDF loss functions are insufficient constraints and fail to ensure convergence to a true distance function, whereas our loss does.

DPS-Net: Deep Polarimetric Stereo Depth Estimation

Paper | Advisor: Zhaopeng Cui

Chaoran Tian, Weihong Pan, Zimo Wang, Mao Mao, Guofeng Zhang, Hujun Bao, Ping Tan, Zhaopeng Cui.
International Conference on Computer Vision. 2023.

tl;dr: We propose a novel deep learning framework called DPS-Net for depth estimation from polarimetric stereo images.

A House Price Valuation Model Based on Geographically Neural Network Weighted Regression

Paper | Advisor: Sensen Wu and Zhenhong Du

Zimo Wang, Yicheng Wang, Sensen Wu, Zhenhong Du.
ISPRS International Journal of Geo-Information 11, no. 8: 450. 2022.

tl;dr: Neural networks can be powerful for learning kernel functions in a spatial weighted regression model to predict house prices.

Hobbies

As a member of the UCSD table tennis team, I don’t mind playing against most people using my phone as a paddle. I always bring my racket🏓 to academic venues. Email me for a friendly match!

I like stargazing. / In the final of the schoolwide freshmen debate competition of Zhejiang University, my team got the second place.

I had my rock&roll band in the high school!

Gallery

After an incredible journey around the world, I'm excited to share some breathtaking views with you. Join me in exploring these beautiful moments captured through my lens!

This website is modified from this source code. No. Visitor Since January 2023.