科研成果

2025
Sun Y, Huang Y, Yang Z, Schneider L-S, Thies M, Gu M, Mei S, Bayer S, Zöllner FG, Maier A. EAGLE: an edge-aware gradient localization enhanced loss for CT image reconstruction. Journal of Medical Imaging. 2025;12:014001–014001.
Yang C, Xie J, Huang X, Tan H, Li Q, Tang Z, Ma X, Lu J, He Q, Fu W, et al. ECS-Net: Extracellular space segmentation with contrastive and shape-aware loss by using cryo-electron microscopy imaging. Expert Systems with Applications. 2025.
Su Y, Li H, Huang* H, Li* D. Effects of the pseudogap and the Fermi surface on the rapid Hall-coefficient changes in cuprates. Phys. Rev. B [Internet]. 2025;111:064518. 访问链接
Lei Z, Shao S, Xiong Y. An efficient stochastic particle method for moderately high-dimensional nonlinear PDEs . Journal of Computational Physics [Internet]. 2025;527:113818. 访问链接Abstract
Numerical resolution of moderately high-dimensional nonlinear PDEs remains a huge challenge due to the curse of dimensionality for the classical numerical methods including finite difference, finite element and spectral methods. Starting from the weak formulation of the Lawson-Euler scheme, this paper proposes a stochastic particle method (SPM) by tracking the deterministic motion, random jump, resampling and reweighting of particles. Real-valued weighted particles are adopted by SPM to approximate the high-dimensional solution, which automatically adjusts the point distribution to intimate the relevant feature of the solution. A piecewise constant reconstruction with virtual uniform grid is employed to evaluate the nonlinear terms, which fully exploits the intrinsic adaptive characteristic of SPM. Combining both, SPM can achieve the goal of adaptive sampling in time. Numerical experiments on the 6-D Allen-Cahn equation and the 7- D Hamiltonian-Jacobi-Bellman equation demonstrate the potential of SPM in solving moderately high-dimensional nonlinear PDEs efficiently while maintaining an acceptable accuracy
Li Y, Tang Y. Embedding Artificial Intelligence into Archival Data Governance: Opportunities, Challenges, and the Chinese Experience. Artificial Intelligence in Records and Information Management. 2025:1-30.
Zhang C, Li W, Luo Z, Zhang P. Engaging with AI in Crowdsourced Digitization of Ancient Texts: User Perception and Interaction, in Annual Meeting of Association for Information Science and Technology.; 2025.
Tang F, Zhang S, Zhu B, Sun J. Enhanced LiDAR Odometry for Autonomous Vehicular Positioning System Using Local Feature Enhancement and Global Motion Constraint. IEEE Transactions on Vehicular Technology. 2025:1-16.
Tang R, Guo H, Gong L, Chen Y, Duan Y, Wang S, Chen Z, Luo F-X, Xiao L. Enhancing the Efficiency of HLCT Emitter via External TTA Up-conversion With Exciton Recycling Channel. ADVANCED OPTICAL MATERIALS. 2025.
Su L, Tang Y. The evolution of archival policies and regulations in China: a topic modelling approach. Archives and Records. 2025:1-19.
Xie J. The existence of Zariski dense orbits for endomorphisms of projective surfaces. With an appendix in collaboration with Thomas Tucker. J. Amer. Math. Soc. [Internet]. 2025;38(1):1-62. pdf
Huang Z, Liang J, Wang Y, Sun Z, Shigekawa N, Li M, WANG R, Cheng Z. Experimental Observation of Extremely Strong Defect-Phonon Scatterings in Semiconductor Single Crystals. arXiv preprint arXiv:2504.20820. 2025.
Lufungulo, E. S. JMJ & K. Exploring factors of open educational resources (OER) in Zambian community schools: A qualitative study. Social Sciences & Humanities Open [Internet]. 2025;11(101465). 访问链接
Lian Y, others. The FAST Globular Cluster Pulsar Survey (GC FANS). Astrophys. J. Suppl. 2025;279:51.
Yu X, Zhong N, Cheng Y, Xin T, Luo Q, Gong T, Chen J, Wu J, Cheng R, Fu Z, et al. Ferroelectric materials, devices, and chips technologies for advanced computing and memory applications: development and challenges. Science China Information Sciences [Internet]. 2025;68:160401. 访问链接
Hou Y, Bert C, Gomaa A, Lahmer G, Höfler D, Weissmann T, Voigt R, Schubert P, Schmitter C, Depardon A, et al. Fine-tuning a local LLaMA-3 large language model for automated privacy-preserving physician letter generation in radiation oncology. Frontiers in Artificial Intelligence. 2025;7:1493716.
Tang X, He P, Zhang Y, Xu Y, Jiang X*. From bench to bucks: measuring the medical technology transfer. The Journal of Technology Transfer [Internet]. 2025:1-24. 访问链接
Ayzenberg D, others. Fundamental physics opportunities with future ground-based mm/sub-mm VLBI arrays. Living Rev. Rel. 2025;28:4.
Liu X-Y, Wang A-Q*, Li D, Zhao T-Y, Liao X, Liao Z-M†. Giant Third-Order Nonlinearity Induced by the Quantum Metric Quadrupole in Few-LayerWTe2. Phys. Rev. Lett. [Internet]. 2025;134(026305). 访问链接
Gao Y, McJeon H, Ou Y, Chen L, Lv J, Fang D, Wang Y, Ye S, Song C, Gao P. Global land system maps at 1 km resolution for 1.5° C climate. Scientific Data [Internet]. 2025;12(1):672. [Link]
Tang F, Zhu B, Sun J. Gradient Enhancement Techniques and Motion Consistency Constraints for Moving Object Segmentation in 3D LiDAR Point Clouds. Remote Sensing [Internet]. 2025;17. 访问链接Abstract
The ability to segment moving objects from three-dimensional (3D) LiDAR scans is critical to advancing autonomous driving technology, facilitating core tasks like localization, collision avoidance, and path planning. In this paper, we introduce a novel deep neural network designed to enhance the performance of 3D LiDAR point cloud moving object segmentation (MOS) through the integration of image gradient information and the principle of motion consistency. Our method processes sequential range images, employing depth pixel difference convolution (DPDC) to improve the efficacy of dilated convolutions, thus boosting spatial information extraction from range images. Additionally, we incorporate Bayesian filtering to impose posterior constraints on predictions, enhancing the accuracy of motion segmentation. To handle the issue of uneven object scales in range images, we develop a novel edge-aware loss function and use a progressive training strategy to further boost performance. Our method is validated on the SemanticKITTI-based LiDAR MOS benchmark, where it significantly outperforms current state-of-the-art (SOTA) methods, all while working directly on two-dimensional (2D) range images without requiring mapping.

Pages