What I cannot create, I do not understand.

– Richard Feynman

I am a PhD student at Stevens Institute of Technology (2021 - now), advised by Prof. Tian Han. My research goal is to develop models that can understand the world in an efficient and effective way, such that my main works focus on probabilistic generative modeling and learning interpretable (hierarchical) data representations in computer vision and machine learning.

My Ph.D research works (until now) can be summarized into

Please don’t hesitate to contact me for collaboration 😊.

Email 📧, Resume 📁, Google Scholar 🎓

Publication

[ICML, 2024] Learning Latent Space Hierarchical EBM Diffusion Models
Jiali Cui, Tian Han

[NeurIPS, 2023] Learning Energy-based Model via Dual-MCMC Teaching
Jiali Cui, Tian Han

[ICCV, 2023] Learning Hierarchical Features with Joint Latent Space Energy-Based Prior
Jiali Cui, Ying Nian Wu, Tian Han

[CVPR, 2023] Learning Joint Latent Space EBM Prior Model for Multi-layer Generator
Jiali Cui, Ying Nian Wu, Tian Han

[NeurIPS, 2020] Semi-supervised Learning by Latent Space Energy-Based Model of Symbol-Vector Coupling
Bo Pang, Erik Nijkamp, Jiali Cui, Tian Han, Ying Nian Wu
Workshop on I Can’t Believe It’s Not Better (ICBINB) @ NeurIPS, 2020

Service

• Reviewer of 2022, ECCV, AAAI
• Reviewer of 2023, ICCV, AAAI, CVPR, NeurIPS
• Reviewer of 2024, AAAI, ICLR, CVPR, ICML, ECCV

Education

• B.S. Harbin Institute of Technology (Harbin, China), 2015 - 2019
• M.S. Stevens Institute of Technology (Hoboken, NJ), 2019 - 2021
• Ph.D Stevens Institute of Technology (Hoboken, NJ), 2021 - Now