Wenxian Shi (施文娴)

I am a researcher at ByteDance AI-Lab MLNLC group, adivsed by Dr. Hao Zhou and Dr. Lei Li. Before that, I received my master's degree from Peking University (PKU), advised by Prof. Junfeng Hu.

Email: wxsh1213 [at] gmail.com / shiwenxian [at] bytedance.com

Github   /   Google Scholar   /   CV     

profile photo
Research Topics

My research interests lie in the general area of machine learning, particularly in representation learning and generative modeling. Currently, I focus on applications on discrete and structured data settings such as natural language processing (e.g., text generation and representation) and computational biology (e.g., molecular generation for drug discovery).

Publication (* equal contribution)

Follow Your Path: a Progressive Method for Knowledge Distillation
Wenxian Shi*,  Yuxuan Song*,  Hao Zhou,  Bohan Li,  Lei Li 
the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD), 2021  

Dispersed Exponential Family Mixture VAEs for Interpretable Text Generation
Wenxian Shi,  Hao Zhou,  Ning Miao,  Lei Li 
Proceedings of the 37th International Conference on Machine learning (ICML), 2020  

Variational Template Machine for Data-to-Text Generation
Rong Ye,  Wenxian Shi,  Hao Zhou,  Zhongyu Wei,  Lei Li 
International Conference on Learning Representations (ICLR), 2020  

Kernelized Bayesian Softmax for Text Generation
Ning Miao,  Hao Zhou,  Chengqi Zhao,  Wenxian Shi,  Lei Li 
the 33rd Conference on Neural Information Processing Systems (NeurIPS), 2019  

Education

  • Sep. 2016 - July 2019: Peking University
    M.S. in Computer Science
    Advisors: Junfeng Hu.
  • Sep. 2012 - July 2016: Peking University
    B.S. in Physics (Yuanpei Program)


  • Updated at Nov. 2020
    Thanks Jon Barron for this amazing work