Huadong Zhang

Towards unified AI-driven fracture mechanics: the extended deep energy method (XDEM)

Yizheng Wang, Yuzhou Lin, Somdatta Goswami, Luyang Zhao et autres

Physics-Informed Neural Networks (PINNs) have recently emerged as powerful tools for solving partial differential equations (PDEs), with the Deep Energy Method (DEM) proving especially effective in fracture mechanics due to its energy-based formulation. Despite these advances, existing DEM approaches require dense collocation …

cn, us, ca, de (code pays fourni par la source)

Plasolver: Physics-Informed Neural Operators for Elastoplasticity

Yizheng Wang, Mohammad Sadegh Eshaghi, Huadong Zhang, Xiaoying Zhuang et autres

Elastoplastic analysis is computationally demanding because its nonlinear, path-dependent constitutive behavior requires incremental loading and repeated iterative solutions. To address this challenge, we propose Plasolver, a physics-informed neural operator framework that combines the efficiency of operator learning with the accuracy and robustness …

Maize yield responses to bio-organic fertilizer N substitution: Stage-specific associations with soil nitrate status, leaf function, and grain filling

Lintao Liu, Hongxia Cao, Haolei Shi, Guoteng Du et autres

cn (code pays fourni par la source)

Correct When Paired, Wrong When Split: Decoupling and Editing Modality-Specific Neurons in MLLMs

Tingchao Fu, Wenkai Wang, Fanxiao Li, Huadong Zhang et autres

Although Knowledge Editing provides an efficient mechanism for updating the knowledge of Multimodal Large Language Models (MLLMs), we find that current paradigms still suffer from an important yet remain underexplored issue : editing decoupling failure, where entity-related knowledge can be updated when …

Correct When Paired, Wrong When Split: Decoupling and Editing Modality-Specific Neurons in MLLMs

Tingchao Fu, Wenkai Wang, Fanxiao Li, Huadong Zhang et autres

Although Knowledge Editing provides an efficient mechanism for updating the knowledge of Multimodal Large Language Models (MLLMs), we find that current paradigms still suffer from an important yet remain underexplored issue : editing decoupling failure, where entity-related knowledge can be updated when …

cn, sg (code pays fourni par la source)

Blockchain-based green power trading strategy for park energy internet

Entang Li, Huadong Zhang, Jixia Zhang, Ming Gao et autres

This study aims to design a blockchain-based green electricity trading mechanism for industrial park energy internet (PEI) systems, enhancing the efficiency and transparency of green electricity transactions. First, the core technologies of blockchain are introduced. Building upon this foundation, the study further …

cn (code pays fourni par la source)

Multiple Local-Guided Modality-Shared Feature Learning Transformer for Visible-Infrared Person Re-Identification

Xiyuan Wang, Huadong Zhang, Shuli Cheng

Visible-Infrared Person Re-Identification (VI-ReID) aims to identify the same person in heterogeneous images captured by different spectral cameras. Due to the different imaging mechanisms of heterogeneous cameras, there are significant modality differences between cross-modal images. Directly extracting modality-shared features from heterogeneous images …

cn (code pays fourni par la source)

Correct When Paired, Wrong When Split: Decoupling and Editing Modality-Specific Neurons in MLLMs

Tingchao Fu, Wenkai Wang, Fanxiao Li, Huadong Zhang et autres

Tingchao Fu, Wenkai Wang, Fanxiao Li, Huadong Zhang, Jinhong Zhang, Dayang Li, Yunyun Dong, Renyang Liu, Wei Zhou. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

cn, sg (code pays fourni par la source)

Dynamic reactive power optimization analysis of new power system based on artificial intelligence algorithm and communication network technology

Weifang Chu, Yangman Li, Weiwei Li, Baoxin Yu et autres

Abstract This paper discusses the reactive power and voltage control via the reactive power optimization method of the power system based on deep reinforcement learning, especially the application of deep deterministic policy gradient (DDPG) algorithm. The study describes the state and dynamic …

cn, in (code pays fourni par la source)

Adaptive Sample Equilibrium and Confidence-Gated Triplet Learning for Robust Unsupervised Person Re-Identification

Tao Liu, Weimin Gao, Y R Li, Huadong Zhang

cn (code pays fourni par la source)

UAV-Assisted Task Offloading for Edge Computing in Blockchain Empowered Smart Grid

Huadong Zhang, Haiqi Li, Entang Li, Fengxue Zhang et autres

To address the challenges of task offloading in mobile edge computing (MEC) for smart grids, this paper proposes an unmanned aerial vehicle (UAV)-enhanced deep reinforcement learning (DRL) approach. By deploying edge services on UAVs to handle dynamic scenarios and integrating blockchain for …

cn (code pays fourni par la source)

Design of Nonlinear Robust Adaptive Optimal Excitation Controller for MFGS

Shaoming Liu, Hanfeng Wang, Shiwei Li, Huadong Zhang et autres

In this paper, based on the robust control theory and the back-stepping methods, the new nonlinear robust control method for methanol-fueled generator set (MFGS) excitation system, which can realize the optimal interference attention, is adaptive to the uncertain parameters and can has …

cn (code pays fourni par la source)