About me

I’m a final-year Ph.D. candidate in Computer Science at the University of Toronto, co-advised by Professors Steve Easterbrook and Animesh Garg. My research focuses on controllable video generation and world models, drawing inspiration from Bayesian Brain Theory. I study how intelligent agents internalize the rules of their environment by predicting sensory signals—especially vision—to enable planning and interaction. My work explores spatiotemporal modeling using deep learning, with applications in video prediction, climate forecasting, and building pretrained models that understand 3D world dynamics for autonomous exploration. I’m currently open to AI/ML roles in both industry and research.

Education

• B.S. in Neuroscience, Nanjing University, 2015
• M.S. in Computational Biology, Harvard University, 2017
• Ph.D. in Computer Science, University of Toronto, 2025

Publications

• EgoSim: Egocentric Exploration in Virtual Worlds with Multi-modal Conditioning.
  Wei Yu, Songheng Yin, Steve Easterbrook, Animesh Garg (ICLR 2025) .
  [PDF] [Project page]

• Modular Action Concept Grounding in Semantic Video Prediction.
  Wei Yu, Wenxin Chen, Songheng Yin, Steve Easterbrook, Animesh Garg (CVPR 2022) .
  [PDF] [Project page]

• Efficient and Information-Preserving Future Frame Prediction and Beyond.
  Wei Yu, Yichao Lu, Steve Easterbrook, Sanja Fidler. 8th International Conference on Learning Representations (ICLR 2020).
  [PDF] [Code]

• Recovering the parameters underlying the Lorenz-96 chaotic dynamics..
  Soukayna Mouatadid., Pierre Gentine, Wei Yu, Steve Easterbrook. (ICML 2019 (workshop)).
  [PDF]

• Application of Deep Learning to Estimate Atmospheric Gravity Wave Parameters in Reanalysis Data Sets..
  D. Matsuoka, S. Watanabe, K. Sato, S. Kawazoe, Wei Yu, Steve Easterbrook. Geophysical Research Letters.
  [PDF]

Experience

• People, AI and Robotics Group, University of Toronto

  Graduate Research Assistant (April 2020 – Present)
  Mentor: Dr. Animesh Garg

• Sustainability Informatics Lab, University of Toronto

  Graduate Research Assistant (August 2018 – Present)
  Mentor: Dr. Steve Easterbrook

• Vector Institute, Toronto

  Student Researcher (April 2020 – Present)

• Snap Research, Los Angeles

  Research Intern (May 2021 – October 2021)
  Mentor: Dr. Sergey Tulyakov

• NVIDIA Toronto AI Lab, Toronto

  Research Intern (May 2019 – August 2019)
  Mentor: Dr. Sanja Fidler

• Cox Lab, Harvard University

  Research Assistant (January 2017 – May 2018)
  Mentor: Dr. David Cox

• CyLab, Carnegie Mellon University (ECE)

  Research Assistant (May 2016 – September 2016)
  Mentor: Dr. Marios Savvides

Award

• 3rd place in Traffic4cast Competition
• Outstanding Paper Award at ICML Workshop on Controllable Video Generation
• Vector Institute Research Grant
• University of Toronto Travel Grant for CVPR and ICLR

Academic Services

• Conference Reviewer: CVPR, NeurIPS, ICLR, ICML, ICCV, ECCV, AAAI