Xiaoyan Wang



Hello! I am a research scientist at Facebook Meta Research. I work on Bayesian modeling and lead the development of the PyTorch-based Bean Machine probabilistic programming language.

Prior to Meta, I received my Master’s degree in computer science at the University of Illinois at Urbana-Champaign (UIUC), where I was advised by Professor Chengxiang Zhai at the TIMAN Group, working on natural language understanding and human-inspired language models. During my undergrad at UIUC, I was forunated to be advised by Professor Sanmi Koyejo and Professor Julia Hockenmaier.


Dec 15, 2021 :sparkles: Bean Machine is officially open-sourced! Checkout our Meta Research blog post for details.
Oct 22, 2020 :sparkles: We just open-sourced PPL Bench (blog post) and presented it at PROBPROG 2020.
May 16, 2020 :mortar_board: I just finished my Master’s degree at UIUC and will be joining Facebook Meta full-time.

selected publications

  1. Improving natural language inference using external knowledge in the science questions domain
    Xiaoyan Wang, Pavan Kapanipathi, Ryan Musa, Mo Yu, Kartik Talamadupula, Ibrahim Abdelaziz, Maria Chang, Achille Fokoue, Bassem Makni, Nicholas Mattei, and Michael Witbrock
    In Proceedings of the AAAI Conference on Artificial Intelligence 2019
  2. Answering science exam questions using query rewriting with background knowledge
    Ryan Musa, Xiaoyan Wang, Achille Fokoue, Nicholas Mattei, Maria Chang, Pavan Kapanipathi, Bassem Makni, Kartik Talamadupula, and Michael Witbrock
    In Automated Knowledge Base Construction 2019
  3. PPL Bench: Evaluation Framework For Probabilistic Programming Languages
    Sourabh Kulkarni, Kinjal Divesh Shah, Nimar Arora, Xiaoyan Wang, Yucen Lily Li, Nazanin Khosravani Tehrani, Michael Tingley, David Noursi, Narjes Torabi, Sepehr Akhavan Masouleh, Eric Lippert, and Erik Meijer
    In International Conference on Probabilistic Programming (PROBPROG) 2020