FUDAN UNIVERSITY · EMBODIED AI
Shengqi Xu 许晟旗
I am a second-year Ph.D. student at the Fudan Vision and Learning Laboratory, Fudan University, advised by Prof. Zuxuan Wu. My research focuses on embodied AI and tactile-driven manipulation.
Previously, I received my master’s degree from Huazhong University of Science and Technology, supervised by Prof. Luxin Yan. I was also fortunate to be mentored by Prof. Wangmeng Zuo. My earlier research focused on imaging through atmospheric turbulence.
News
- 2026.09 Our work ThinkingVLA for Interleaved Chain-of-Thought VLA has been accepted by CoRL’26!
- 2026.07 We introduce 𝒩0-Foundation, a tactile-centric foundation for embodied manipulation spanning infrastructure, data, representation learning, and benchmarks.
- 2026.07 We introduce 𝒩0-VTLA, a vision-tactile-language-action foundation model with latent tactile tokens.
- 2026.07 We introduce 𝒩0-TWAM, a tactile-native world-action model for contact-rich manipulation.
- 2026.06 Our work ViTacMotor for Visuo-Tactile Manipulation with Tactile Motion Correlation has been accepted by ECCV’26!
- 2026.01 Our work PreferThinker for reasoning-based personalized image preference assessment has been accepted by ICLR’26!
- 2024.07 Our work CDSP for long-range turbulence mitigation with a large-scale benchmark has been accepted by ECCV’24!
- 2024.06 We have won 1st place in the track ‘Atmospheric Turbulence Mitigation’ in the CVPR’24 7th UG2+ Challenge!
- 2023.06 We have won 1st place in the track ‘Atmospheric Turbulence Mitigation’ in the CVPR’23 6th UG2+ Challenge!
Research
* Corresponding author.
Tactile-driven Manipulation
GuidedVTLA: Where Proactive Tactile Guidance Meets Reactive Tactile Refinement
Shengqi Xu, Guojin Zhong, Yutao Fan, Tianyi Lu, Yuhan Zhu, Yutao Ling, Zhiying Du, Shunlin Lu, Ziyi Ye, Zuxuan Wu*, Yu-Gang Jiang*
Under Review, 2026.
OpenViTac: Learning and Benchmarking Visuo-Tactile Policies in a Unified Sim-and-Real Framework
Yifan Wu, Qin Li, Nan Min, Guojin Zhong, Haoyu Zhao, Zhiyuan Li, Houze Xu, Shengqi Xu, Xingyao Lin, Zijie Diao, Zhaoxiang Liu, Shiguo Lian, Shunlin Lu, Shihao Zhao, Ziyi Ye, Zuxuan Wu, Yu-Gang Jiang
arXiv preprint, 2026.
Shengqi Xu, Guojin Zhong, Yang Liu, Fanjie Wang, Hu Luo, Hanyu Zhou, Weiyao Zhang, Ziyi Ye, Zuxuan Wu*, Yu-Gang Jiang*
European Conference on Computer Vision ( ECCV ), 2026.
𝒩0-Foundation: Towards the Age of Tactile Intelligence
NeoteAI Team and Fudan TEAI Team
Core Contributor
A tactile-centric foundation that brings together sensing hardware, large-scale multimodal data, transferable contact representations, and standardized benchmarks. NeoData and NeoForce connect scalable interaction data with hardware-agnostic tactile learning.
Technical Report , 2026.
𝒩0-VTLA: Scaling Vision-Tactile-Language-Action Model with Latent Tactile Tokens
NeoteAI Team and Fudan TEAI Team
Core Contributor
A vision-tactile-language-action foundation model pretrained on NeoData for contact-rich manipulation. Predictive tactile tokens bring learned contact dynamics into action generation, enabling policies to use touch alongside vision and language.
Technical Report , 2026.
𝒩0−TWAM: Scaling Tactile-Native World Action Model for Contact-Rich Manipulation
NeoteAI Team and Fudan TEAI Team
Core Contributor
A tactile-native world-action model that jointly predicts future visual observations and contact. Large-scale visuo-tactile pretraining across six embodiments and 450 tasks supports action learning for contact-rich manipulation.
Technical Report , 2026.
Vision Language Action Model
ThinkingVLA: Interleaved Vision and Language Reasoning for Robotic Manipulation
Tianyi Lu, Hui Zhang, Zijie Diao, Junke Wang, Shengqi Xu, Xingyao Lin, Guojin Zhong, Ziyi Ye, Peng Wang, Zuxuan Wu, Yu-Gang Jiang
Conference on Robot Learning (CoRL), 2026.
Vision Language Model
PreferThinker: Reasoning-based Personalized Image Preference Assessment
Shengqi Xu, Xinpeng Zhou, Yabo Zhang, Ming Liu*, Tao Liang, Tianyu Zhang, Yalong Bai, Zuxuan Wu, Wangmeng Zuo
International Conference on Learning Representations ( ICLR ), 2026.
Computational Photography
Long-range Turbulence Mitigation: A Large-scale Dataset and A Coarse-to-fine Framework
Shengqi Xu, Run Sun, Yi Chang*, Shuning Cao, Xueyao Xiao, Luxin Yan
European Conference on Computer Vision ( ECCV ), 2024.
Shengqi Xu, Xueyao Xiao, Shuning Cao, Yi Chang*, Luxin Yan
6th CVPRW UG2+ Challenge, 2023.
First Place in the Track of Text Recognition through Turbulence
Shengqi Xu, Shuning Cao, Haoyue Liu, Xueyao Xiao, Yi Chang*, Luxin Yan
6th CVPRW UG2+ Challenge, 2023.
First Place in the Track of Code Recognition through Turbulence
Honors and Awards
- 2025.04 Outstanding Graduates of HUST.
- 2024.10 National Scholarship (Highest Honor: Top 2% Nationwide).
- 2024.06 First Place of the track ‘Atmospheric Turbulence Mitigation’ in the CVPR 2024 UG2+ Challenge.
- 2023.06 First Place of the track ‘Atmospheric Turbulence Mitigation’ in the CVPR 2023 UG2+ Challenge.
- 2023.10 Outstanding Student Paper Award at the 6th Conference on Atmospheric and Adaptive Optics.
- 2020.04 First Prize of the Asia and Pacific Mathematical Contest in Modeling.
- 2020.04 First Prize of Chinese Mathematics Competition.
Education
- 2025.09 - Now,
Fudan University, China. Ph.D. Candidate - 2022.09 - 2025.06,
Huazhong University of Science and Technology, China.
Master of Engineering - 2018.09 - 2022.06,
Zhengzhou University, China.
Bachelor of Engineering
Ranking: 4/88
Professional Service
Journal Reviewer
- IEEE Transactions on Computational Imaging (TCI)
- IEEE Transactions on Circuits and Systems for Video Technology (T-CSVT)
Conference Reviewer
- CVPR 24
- ECCV 24
- ACMMM 24