About
I am a second-year Ph.D. student in the Department of Civil Engineering at Tsinghua University, advised by Prof. Jiansheng Fan. I received my B.Eng. in Civil Engineering from Tsinghua University in 2024.
My current research interests include spatial intelligence and physical AI. Previously, I worked on spatio-temporal data mining and structural optimization.
News
Publications
* equal contribution; † Corresponding author.
Thinking in Structures: Evaluating Spatial Intelligence in Constraint-Governed Spaces
Chen Yang, Guanxin Lin, Youquan He, Peiyao Chen, Guanghe Liu, Yufan Mo, Zhouyuan Xu, Linhao Wang, Guohui Zhang, Zihang Zhang, Shenxiang Zeng, Chen Wang†, Jiansheng Fan†
ICML 2026CCFA
Introduced SSI-Bench, constructed from complex real-world 3D structures with feasible configurations tightly governed by geometric, topological, and physical constraints.
Towards Reliable Deep Excavation Monitoring through Graph Recurrent Neural Network-Based Spatio-Temporal Imputation
Chen Yang, Xiao-guang Zhang, Guo-hui Zhang, Chen Wang†, Jian-sheng Fan
Reliability Engineering & System Safety, 2025IF 11.0
Proposes a GNN + BiGRU spatio-temporal imputation model tailored to excavation monitoring, achieving accurate missing-data recovery and improving monitoring reliability under strong spatial but weak temporal correlations.
DRIK: Distribution-Robust Inductive Kriging without Information Leakage
Chen Yang*, Changhao Zhao*, Chen Wang†, Jiansheng Fan
arXiv, 2025
Introduces DRIK, a distribution-robust inductive Kriging framework that explicitly avoids information leakage and improves interpolation robustness under distribution shift.
Towards Efficient Structural Inverse Analysis Based on AI-Driven Differentiable Optimization Method
Chen Wang, Chong Zhang, Chen Yang†, Jian-sheng Fan
Mechanical Systems and Signal Processing, 2025IF 8.9
Develops an AI-driven differentiable optimization framework for structural inverse analysis, reducing end-to-end inversion time to about one-tenth of conventional pipelines.
Smart Virtual Sensing for Deep Excavations Using Real-Time Ensemble Graph Neural Networks
Chen Yang, Chen Wang†, Feng Zhao, Bin Wu, Jian-sheng Fan, Yu Zhang
Automation in Construction, 2025IF 11.5
Introduces virtual sensing for deep excavation monitoring and proposes a real-time ensemble GNN model for field-scale settlement estimation with improved accuracy and robustness.
Settlement Estimation during Foundation Excavation Using Pattern Analysis and Explainable AI Modeling
Chen Yang, Chen Wang†, Bin Wu, Feng Zhao, Jian-sheng Fan, Lu Zhou
Automation in Construction, 2024IF 11.5
Develops spatio-temporal pattern mining for excavation data to reveal strong spatial dependence, and further builds an explainable spatio-temporal deep learning model for high-accuracy settlement estimation.
Differentiable Automatic Structural Optimization Using Graph Deep Learning
Chong Zhang, Mu-xuan Tao, Chen Wang†, Chen Yang, Jian-sheng Fan
Advanced Engineering Informatics, 2024IF 9.9
Presents a surrogate-model + differentiable optimization framework for structural optimization, enabling end-to-end gradients and achieving approximately 25,000× speedup over traditional heuristic FEM-based methods.
Automatic Design Method of Building Pipeline Layout Based on Deep Reinforcement Learning
Chen Yang, Zhe Zheng, Jia-rui Lin†
The 30th EG-ICE: International Conference on Intelligent Computing in Engineering, 2023
Introduces deep reinforcement learning for automatic building pipeline layout design, improving planning efficiency compared with traditional heuristic algorithms.