I am a third-year PhD student (2023-) in the Department of Computer Science at University of Illinois Chicago (UIC), supervised by Prof. Yan Yan. Prior to joining UIC, I spent a year at Illinois Institute of Technology. I hold both a bachelor’s (IEEE honor class) and a master’s degree from Shanghai Jiao Tong University, where I was fortunate to be advised by Prof. Junchi Yan. Additionally, I had a wonderful time working as a research intern with Prof. Anqi Liu and Prof. Anima Anandkumar at Caltech. My research interests include Machine Learning Efficiency, 3D Vision and Robotics. Most of the publications can be accessed here.
I am currently a visiting student at the UCLA VAIL lab. I was also a Research Intern at Cisco Research, where I was fortunate to work with Gaowen Liu and Ramana Kompella.
🔥 News
- 2026.02: AutoHorizon, the first test-time method for determining the execution horizon for flow-based VLAs, is released!
- 2026.01: REMAC is accepted to ICLR 2026!
- 2025.09: 3 co-authored papers accepted to NeurIPS 2025!
- 2025.06: QuEST and CaO2 are accepted to ICCV 2025!
- 2025.05: Working as a research intern at Cisco Research
- 2025.03: LTDD is accepted to CVPR 2025!
- 2024.11: Serving as the web co-chair for ICMR 2025
- 2024.06: PTQ4DiT is accepted to NeurIPS 2024!
📝 Selected Publications
Real-Time Robot Execution with Masked Action Chunking
Haoxuan Wang, Gengyu Zhang, Yan Yan, Yuzhang Shang, Ramana Rao Kompella, Gaowen Liu
- A real-time robot execution strategy for asynchronous inference.
Haoxuan Wang, Gengyu Zhang, Yan Yan, Ramana Rao Kompella, Gaowen Liu
- The first test-time method for dynamically and automatically determining the execution horizon for flow-based VLAs
QuEST: Low-bit Diffusion Model Quantization via Efficient Selective Finetuning
Haoxuan Wang, Yuzhang Shang, Zhihang Yuan, Junyi Wu, Junchi Yan, Yan Yan
- Parameter efficient finetuning method for diffusion model quantization.
CaO$_2$: Rectifying Inconsistencies in Diffusion-Based Dataset Distillation
Haoxuan Wang, Zhenghao Zhao, Junyi Wu, Yuzhang Shang, Gaowen Liu, Yan Yan
- Diffusion based method for efficient dataset distillation.
Distilling Long-tailed Datasets
Zhenghao Zhao*, Haoxuan Wang*, Yuzhang Shang, Kai Wang, Yan Yan
- Pioneering work confronting biased dataset distillation.
PTQ4DiT: Post-training Quantization for Diffusion Transformers
Junyi Wu*, Haoxuan Wang*, Yuzhang Shang, Mubarak Shah, Yan Yan
- Pioneering work for DiT quantization.
Learning Calibrated Uncertainties for Domain Shift: A Distributionally Robust Learning Approach
Haoxuan Wang, Zhiding Yu, Yisong Yue, Animashree Anandkumar, Anqi Liu and Junchi Yan
- A novel framework for learning calibrated uncertainties under domain shifts.
-
[NeurIPS 2025 (Spotlight)] X-Field: A Physically Grounded Representation for 3D X-ray Reconstruction, Feiran Wang*, Jiachen Tao*, Junyi Wu*, Haoxuan Wang, Bin Duan, Kai Wang, Zongxin Yang, Yan Yan
-
[NeurIPS 2025] Efficient Multimodal Dataset Distillation via Generative Models, Zhenghao Zhao, Haoxuan Wang, Junyi Wu, Yuzhang Shang, Gaowen Liu, Yan Yan
-
[NeurIPS 2025] Orientation-anchored Hyper-Gaussian for 4D Reconstruction from Casual Videos, Junyi Wu, Jiachen Tao, Haoxuan Wang, Gaowen Liu, Ramana Rao Kompella, Yan Yan
-
[Preprint] Leveraging Angular Information Between Feature and Classifier for Long-tailed Learning: A Prediction Reformulation Approach, Haoxuan Wang, Junchi Yan
🎖 Honors and Awards
- 2021-22 First Award of SJTU scholarship
- 2020.06 Graduation with honor, University Graduate Excellence Award of SJTU
- 2019.11 First Award of Zhiyuan Research Program
📖 Education
- 2024.12 - present PhD, University of Illinois at Chicago. (Transferred from IIT)
- 2023.09 - 2024.12 PhD, Illinois Institute of Technology.
- 2020.09 - 2023.03 Master, Shanghai Jiao Tong University.
- 2016.09 - 2020.06 Undergraduate, IEEE Honor Class, Shanghai Jiao Tong University.
💻 Internships
- 2025.05 - 2025.08, Research Intern, Cisco Research.
- 2021 - 2022, Research Intern, Caltech.
