avatar

Cai Zhou 周偲

2nd-year PhD @ MIT EECS.
Massachusetts Institute of Technology
caiz428@mit.edu


About Me

I am a second year PhD student in MIT EECS, co-advised by Tommi Jaakkola and Stephen Bates. I am affiliated with CSAIL and LIDS. My recent research focuses on the intersection of machine learning, generative models and AI4Science. In the summer of 2025, I am interning at Microsoft Research working with Dinghuai Zhang.

Prior to my PhD, I received a bachelor from the Department of Automation, School of Information at Tsinghua University. I am also a member of Class of General Artificial Intelligence of Tsinghua University, and I minor in statistics. In my undergraduate research, I am fortunate to work with Muhan Zhang at Institute for Artificial Intelligence of Peking University. We have great collaborations working on machine learning theories, graph neural networks, AI4Science and LLMs/foundation models. In the summer of 2023, I had the privilege to intern at UCSD advised by Yusu Wang and Rose Yu on machine learning. I also worked with Gao Huang at Tsinghua University on computer vision.

Research Interests

My research interest lies broadly in theoretical and applied machine learning. I aim to understand the foundations of machine learning, with a special focus on its probabilistic and geometric nature. I’m also interested in application areas including computer vision, natural language processing and computational biology. Some of my interested research topics include:

News

Publications & Preprints

  1. Preprint
    Coevolutionary Continuous Discrete Diffusion：Make Your Diffusion Language Model a Latent Reasoner
    Cai Zhou, Chenxiao Yang, Yi Hu, Chenyu Wang, Chubin Zhang, Muhan Zhang, Lester Mackey, Tommi Jaakkola, Stephen Bates, Dinghuai Zhang
    Preprint.
    PDF

  2. Preprint
    What Affects the Effective Depth of Large Language Models?
    Yi Hu, Cai Zhou, Muhan Zhang
    Preprint. Also in NeurIPS 2025 Mechanistic Interpretability Workshop.
    PDF

  3. Preprint
    CLAP：Contrastive Latent Action Pretraining for Learning Vision-Language-Action Models from Human Videos
    Chubin Zhang, Jianan Wang, Zifeng Gao, Yue Su, Tianru Dai, Cai Zhou, Jiwen Lu, Yansong Tang
    Preprint.
    PDF

  4. ICLR 2026
    On Powerful Ways to Generate：Autoregression, Diffusion, and Beyond
    Chenxiao Yang, Cai Zhou, David Wipf, Zhiyuan Li
    Fourteenth International Conference on Learning Representations (ICLR), 2026.
    PDF

  5. ICLR 2026
    SPG：Sandwiched Policy Gradient for Masked Diffusion Language Models
    Chenyu Wang, Paria Rashidinejad, DiJia Su, Song Jiang, Sid Wang, Siyan Zhao, Cai Zhou, Shannon Zejiang Shen, Feiyu Chen, Tommi Jaakkola, Yuandong Tian, Bo Liu
    Fourteenth International Conference on Learning Representations (ICLR), 2026.
    PDF

  6. NeurIPS 2025
    Next Semantic Scale Prediction via Hierarchical Diffusion Language Models
    Cai Zhou*, Chenyu Wang*, Dinghuai Zhang*, Shangyuan Tong, Yifei Wang, Stephen Bates†, Tommi Jaakkola†
    Thirty-ninth Conference on Neural Information Processing Systems (NeurIPS), 2025.
    PDF Code

  7. NeurIPS 2025
    Learning Diffusion Models with Flexible Representation Guidance
    Chenyu Wang*, Cai Zhou*, Sharut Gupta, Zongyu Lin, Stefanie Jegelka, Stephen Bates, Tommi Jaakkola
    Thirty-ninth Conference on Neural Information Processing Systems (NeurIPS), 2025. Also Oral in ICML 2025 FM4LS Workshop, ICML 2025 GenBio Workshop.
    PDF Code

  8. EMNLP 2025 Oral
    Thought Calibration：Efficient and Confident Test-Time Scaling
    Menghua Wu, Cai Zhou, Stephen Bates, Tommi Jaakkola
    The 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP), main. Oral.
    PDF

  9. ICML 2025 Spotlight
    Geometric Representation Condition Improves Equivariant Molecule Generation
    Zian Li*, Cai Zhou*, Xiyuan Wang, Xingang Peng, Muhan Zhang
    Forty-second International Conference on Machine Learning (ICML), 2025. Spotlight.
    PDF Code

  10. Preprint
    Towards Stable, Globally Expressive Graph Representations with Laplacian Eigenvectors
    Junru Zhou, Cai Zhou, Xiyuan Wang, Pan Li, Muhan Zhang
    Preprint.
    PDF

  11. Preprint
    FiTv2：Scalable and Improved Flexible Vision Transformer for Diffusion Model
    Zidong Wang, Zeyu Lu, Di Huang, Cai Zhou, Wanli Ouyang, Lei Bai
    Preprint.
    PDF Code

  12. AAAI 2025
    ChemVLM：Exploring the Power of Multimodal Large Language Models in Chemistry Area
    ..., Cai Zhou, ...
    Thirty-Ninth AAAI Conference on Artificial Intelligence (AAAI), 2025.
    PDF

  13. NeurIPS 2024
    Unifying Generation and Prediction on Graphs with Latent Graph Diffusion
    Cai Zhou, Xiyuan Wang, Muhan Zhang
    Thirty-eighth Conference on Neural Information Processing Systems (NeurIPS), 2024.
    PDF Code BibTex

  14. IJCV 2024
    Locally Supervised Deep Learning by Maximizing Information Propagation
    Yulin Wang, Zanlin Ni, Yifan Pu, Cai Zhou, Jixuan Ying, Shiji Song, Gao Huang
    International Journal of Computer Vision (IJCV), 2024.
    PDF

  15. AISTATS 2024
    On the Theoretical Expressive Power and Design Space of High Order Graph Transformers
    Cai Zhou, Rose Yu, Yusu Wang
    Twenty-seventh International Conference on Artificial Intelligence and Statistics (AISTATS), 2024.
    PDF Code BibTex

  16. NeurIPS 2023
    Facilitating Graph Neural Networks with Random Walk on Simplicial Complexes
    Cai Zhou, Xiyuan Wang, Muhan Zhang
    Thirty-seventh Conference on Neural Information Processing Systems (NeurIPS), 2023.
    PDF Code BibTex

  17. ICML 2023
    From Relational Pooling to Subgraph GNNs：A Universal Framework for More Expressive Graph Neural Networks
    Cai Zhou*, Xiyuan Wang*, Muhan Zhang
    Fortieth International Conference on Machine Learning (ICML), 2023.
    PDF Code BibTex

Services

Conference Reviewers


Powered by Jekyll and Minimal Light theme.