My name is Zexuan Zhong. I am a M.S. student at Department of Computer Science, University of Illinois at Urbana-Champaign. I am now working with Prof. Tao Xie on applying machine learning to address software engineering problems. I also work closely with Prof. Jian Peng. Before coming to UIUC, I received a B.S. in Computer Science from Peking University. During summer 2018, I was doing research at MIT Media Lab under supervision of Prof. Fadel Adib. I was also a full-time research intern at Microsoft Research Asia in my undergrad senior year.
Besides software engineering and machine learning research, I also am very interested in intersection between computer vision and computer graphics with topics such as computational photography, digital image editing, image-based rendering, and 3D modeling.


  1. SemRegex: A Semantics-Based Approach for Generating Regular Expressions from Natural Language Specifications
    Zexuan Zhong, Jiaqi Guo, Wei Yang, Jian Peng, Tao Xie, Jian-Guang Lou, Ting Liu, Dongmei Zhang
    EMNLP 2018 (appear soon)
  2. CoLink: An Unsupervised Framework for User Identity Linkage
    Zexuan Zhong, Yong Cao, Mu Guo, Zaiqing Nie
    Download: [PDF]
  3. Generating Regular Expressions from Natural Language Specifications: Are We There Yet?
    Zexuan Zhong, Jiaqi Guo, Wei Yang, Tao Xie, Jian-Guang Lou, Ting Liu, Dongmei Zhang
    Download: [PDF]

Selected Projects

Regexs generation from NL sentences (ongoing)
We focus on automatically generating regular expressions from natural language specifications. Existing approaches reduce the target problem to a general machine translation task. We find a huge drop of performance when applying existing approaches on real-world data vs synthetic data. We propose to utilize string examples of regexs to define the semantic correctness, and use reinforcement learning to maximize the semantic correctness of regexs predicted by the model.
User Identity Linkage
User Identity Linkage refers to match identical users across two social networks. In this project, we propose an unsupervised framework based on co-training to address this problem. We develop an attribute-based model and a relationship-based model. The models will be enhanced in iterative way via co-training. We also propose to employ sequence-to-sequence learning to measure the similarity of attributes, which is an effective implementation of attribute-based model.
DualFocus is a tool for image refocusing. DualFocus takes two images with different focal lengths as input, and outputs an all-in-focus image. For each image, we solve a deconvolution problem with constrains based on prior characteristics of natural in-focus images. We then use Alternating Direction Method of Multipliers (ADMM) to minimize the differences between the deconvolution results for two input images. DualFocus does not require modification of cameras (i.e., coded aperture), and outperformances single-image-based refocusing approaches.
ObjRender is a tool to render 3D synthetic objects in images. We use a mirror ball as a light probe to obtain environment mapping. We take several low dynamic range (LDR) images of the mirror ball to recover a high dynamic range (HDR) image. We build the environment lighting box by mapping the HDR image in sphere domain to the equirectangular domain. We annotate necessary geometry models such as tables in the scene, and render the synthetic objects on those models.
Image Inpainting
Given an image with a hole, we synthesize texture from neighbor regions and fill the texture into the hold. To find the visually plausible texture to be used to fill, we select the sample with similar overlapping region. To preserve edges of objects in the image, we define a priority function based on gradient magnitude. With the priority function, the pixels in edges will be filled first.
Water Simulation
In this project, we simulate the surface of the water in a pool. To simulate the ripple on the water, we use normal mapping techniques in illumination calculation, i.e., we modify the normal vector of each point of the 3D models so that the illumination will relect as ripple, instead of directly simulating the ripple in modelling level.


  • 2017.8 - present, University of Illinois at Urbana-Champaign
    M.S. in Computer Science
  • 2013.9 - 2017.7, Peking University
    B.S. in Computer Science and Technology


  • 2017.8 - present, Research Assistant, University of Illinois at Urbana-Champaign
    Advisor: Prof. Tao Xie
  • 2018.5 - 2018.8, Visiting Research Assistant, MIT Media Lab
    Advisor: Prof. Fadel Adib
  • 2016.7 - 2017.5, Research Intern, Microsoft Research Asia
    Mentors: Dr. Zaiqing Nie and Dr. Yong Cao
  • 2015.11 - 2016.5, Research Assistant, Peking University
    Advisor: Prof. Yingfei Xiong


  • Tang Li-Xin Scholar (4 in 400), 2014 - present
  • Beijing City Excellent Graduate, 2017
  • Peking University Excellent Graduate, 2017
  • Bachelor of Science Summa Cum Laude, 2017
  • Peking University Merit Student, 2014 - 2017

Program Contests

  • Gold Medal, ACM-ICPC Mid-Central USA Regionals Chicago site, 2017
  • Gold Medal, ACM-ICPC Asia Regionals Changchun site, 2013
  • First Prize, China National Olympiad in Informatics, 2012


3107 Siebel Center
201 N. Goodwin Ave.
Urbana, IL 61801, USA
Email: zexuan2@illinois.edu