Minghao Fu

Minghao Fu Minghao Fu Chinese name

Ph.D. student, Halıcıoğlu Data Science Institute, UC San Diego

I am a Ph.D. student at the Halıcıoğlu Data Science Institute, UC San Diego, advised by Biwei Huang.

Before joining UCSD, I was an M.S. student at MBZUAI and a visiting student at CMU, working with Kun Zhang, Peter Spirtes, and Yingyao Hu. I worked on representation learning and generative models, researching how to make them interpretable and controllable. Prior to MBZUAI, I obtained my Bachelor from UESTC, and spend time at Shanghai AI Lab.

Email GitHub Google Scholar LinkedIn Twitter Curriculum Vitae

Research

My research centers on inverting observation-generation to unveil the hidden world. I explore: (1) how to identify a hidden world based on the assumed generative mechanism, and (2) how such a hidden world instructs the observed world evolving in response to the environment.

Hence, my works cover world models (multi-modal learning, VLMs, temporal state-space models), decision-making (VLAs, reinforcement learning), and theoretical aspects of generative models.

News

  • Jan 2026 Two papers on Reinforcement Learning and Multi-modality were accepted to ICLR 2026.
  • Dec 2025 Attended NeurIPS 2025 in San Diego, CA, USA.
  • Nov 2025 Presented our work at The 5th Measurement Errors and Latent Variables Workshop at JHU. Thanks to Dean. Hu for invitation.
  • Sep 2025 Two papers on Temporal State-Space Models were accepted to NeurIPS 2025.

Selected Publications

Ada-Diffuser: Latent-Aware Adaptive Diffusion for Decision-Making

Fan Feng, Selena Ge, Minghao Fu, Zijian Li, Yujia Zheng, Zeyu Tang, Yingyao Hu, Biwei Huang, Kun Zhang
ICLR 2026
arXiv

PersonaX: Multimodal Datasets with LLM-Inferred Behavior Traits

Loka Li*, Wong Yu Kang*, Minghao Fu, Guangyi Chen, Zhenhao Chen, Gongxu Luo, Yuewen Sun, Salman Khan, Peter Spirtes, Kun Zhang
ICLR 2026
arXiv • Project Page • Code • Dataset

Towards Identifiability of Hierarchical Temporal Causal Representation Learning

Zijian Li*, Minghao Fu*, Junxian Huang, Yifan Shen, Ruichu Cai, Yuewen Sun, Guangyi Chen, Kun Zhang
NeurIPS 2025
arXiv • Code

Online Time Series Forecasting with Theoretical Guarantees

Zijian Li, Changze Zhou, Minghao Fu, Sanjay Manjunath, Fan Feng, Guangyi Chen, Yingyao Hu, Ruichu Cai, Kun Zhang
NeurIPS 2025
arXiv

Learning General Causal Structures with Hidden Dynamic Process for Climate Analysis

Minghao Fu, Biwei Huang, Zijian Li, Yujia Zheng, Ignavier Ng, Guangyi Chen, Yingyao Hu†, Kun Zhang†
arXiv

Education

University of California, San Diego Aug 2025 – Present
Ph.D., Halıcıoğlu Data Science Institute
Advisor: Biwei Huang
Carnegie Mellon University Jun 2024 – Nov 2024
Visiting Student, CLeaR
Advisors: Peter Spirtes, Kun Zhang
Mohamed bin Zayed University of Artificial Intelligence Aug 2023 – Jun 2025
M.S., Machine Learning
Advisor: Kun Zhang
University of Electronic Science and Technology of China Sep 2019 – Jun 2023
B.S., Software Engineering
Advisor: Jie Shao

Service

Conference Reviewer: ICML 2026; CVPR 2026; WMW 2026; ICLR 2026; NeurIPS 2025; BMVC 2023.

Journal Reviewer: IEEE Transactions on Image Processing (TIP) 2024