检索结果-内蒙古大学图书馆

5th International Symposium on Artificial Intelligence for Medicine Science, ISAIMS 2024

作者： Chen, Yulin Luo, Jing Tu, Xinhui College of Computer Science and Technology Wuhan University of Science and Technology Hubei Wuhan China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Hubei Wuhan China College of Computer Science Central China Normal University Hubei Wuhan China

ISBN: (纸本)9798400717826

Recent advancements in large language models have led to significant improvements in various natural language processing tasks, including automated question answering. However, these models still struggle with providing accurate responses to complex biology questions, such as those found in the National College Entrance Examination. To address this issue, researchers have explored retrieval-augmented generation techniques, which incorporate external knowledge sources to enhance answer accuracy. While these methods have shown promise in improving efficiency, they often lack the ability to perform in-depth analysis of complex questions. In response to these limitations, the paper proposes a novel ReAct agent-based question answering method. This method combines biology textbook document libraries with Google search engine capabilities, leveraging the strengths of large language models. By integrating these components, this method can engage in autonomous decision-making and multi-step reasoning, allowing for a more thorough analysis of complex biology questions. Experimental results demonstrate the effectiveness of this new approach. When compared to responses generated directly by ChatGPT and GPT-4o models, the method showed significant improvements in accuracy for biology questions from the National College Entrance Examination. Specifically, the new method achieved accuracy increases of 11.1% and 4.64% over ChatGPT and GPT-4o, respectively. © 2024 Copyright held by the owner/author(s).

关键词： Question answering

来源：评论

学校读者我要写书评

暂无评论

S2KAN-SLAM: Elastic Neural LiDAR SLAM with SDF Submaps and Kolmogorov-Arnold Networks

引用

IEEE Transactions on Circuits and systems for Video Technology 2025年

作者： Wang, Zhong Zhang, Lin Wang, Hesheng Shanghai Jiao Tong University State Key Laboratory of Avionics Integration and Aviation System-of-Systems Synthesis Department of Automation Key Laboratory of System Control and Information Processing of Ministry of Education Shanghai200240 China Tongji University School of Computer Science and Technology National Pilot Software Engineering School with Chinese Characteristics Shanghai201804 China

Traditional LiDAR SLAM approaches prioritize localization over mapping, yet high-precision dense maps are essential for numerous applications involving intelligent agents. Recent advancements have introduced methods leveraging neural fields to enhance mapping capabilities;however, these approaches still face several limitations. Firstly, concerning scene representation, they typically employ neural fields with high-dimensional features and multi-layer perceptron decoders utilizing non-continuous activation functions. This results in low learning efficiency and challenges in capturing high-frequency signals. Secondly, in terms of scene organization, these methods often treat the entire scene as a singular neural field, leading to inefficiencies, inflexibility, and difficulties in rectifying accumulated errors when mapping large-scale environments over extended periods. To tackle the first issue, we propose a lightweight continuous SDF regression approach by encoding the scene in single-valued embeddings and decoding SDF values from a Kolmogorov-Arnold Network. By minimizing discrepancies in measuring range, sampling distance, and decoded SDF values, we facilitate iterative frame-to-model tracking and bundle adjustment neural mapping. To mitigate the second challenge, we propose structuring the whole scene into multiple neural SDF submaps. By establishing node-node, node-submap, and loop closure constraints into a global pose graph, the system can create dense neural maps with global consistency across large-scale scenes. Experimental evaluations in both real-world and simulated settings indicate that our system achieves superior mapping completeness and accuracy, enhanced learning efficiency, reduced memory consumption, and greater flexibility compared to its counterparts. © 2025 IEEE.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

DD-Net: Dynamic Network Architecture for Optimized Curve Segmentation and Reduce Computational Redundancy 27th

DD-Net: Dynamic Network Architecture for Optimized Curve Seg...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Cao, Yunxiang Chen, Li Wang, Yubo Feng, Zhida Tian, Jing School of Computer Science and Technology Wuhan University of Science and Technology Wuhan China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan University of Science and Technology Wuhan China Institute of Systems Science National University of Singapore Singapore Singapore

ISBN: (纸本)9783031783821

Curvilinear Structure segmentation has numerous applications in various fields, including providing a better understanding of defects such as cracks on roads or walls, thereby assessing the structural safety of buildings. Numerous curve segmentation methods based on Convolutional Neural Networks(CNN) have been developed in recent years. However, preserving feature integrity remains a challenge. To address this issue, this paper introduces a double dynamic network, called DD-Net, which consists of a dynamic structure for training and dynamic masking for inference. Firstly, to enhance the nonlinearity of the model, a paradigm to auto-adjust the CNN structure via a trainable module is formed for better fine-grained feature extraction. This dynamic structure (DS) paradigm enables a trade-off to extract or discard the feature data in the model training. Due to the data heterogeneity between the training and inference data, models may contain some useless nodes that are less effective during inference. The dynamic masking (DM) is proposed to omit the useless nodes based on the score difference between the train and inference feature statistics, thereby reducing redundant computations. To further improve the model’s performance on thin-curve segmentation and to preserve feature integrity, we introduce a non-curve suppression (NCS) module. This module focuses on background information while considering foreground prediction to address noisy conditions. The experimental results show that our DD-Net achieves promising results on three benchmark datasets and outperforms state-of-the-art curve segmentation models. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

UW-DNeRF: Deformable Soft Tissue Reconstruction with Uncertainty-Guided Depth Supervision and Local information Integration

引用

IEEE Transactions on Medical Imaging 2025年 PP卷 PP页

作者： Shan, Jiwei Zhang, Zixin Li, Hao Hsieh, Cheng-Tai Li, Yirui Wu, Wenhua Wang, Hesheng School of Electronic Information and Electrical Engineering of Shanghai Jiao Tong University Shanghai200240 China Shanghai Jiao Tong University School of Mechanical Engineering Shanghai200240 China Shanghai Jiao Tong University State Key Laboratory of Avionics Integration and Aviation System-of-Systems Synthesis Department of Automation Key Laboratory of System Control and Information Processing of Ministry of Education Shanghai200240 China

Reconstructing deformable soft tissues from endoscopic videos is a critical yet challenging task. Leveraging depth priors, deformable implicit neural representations have seen significant advancements in this field. However, depth priors from pre-Trained depth estimation models are often coarse, and inaccurate depth supervision can severely impair the performance of these neural networks. Moreover, existing methods overlook local similarities in input sequences, which restricts their effectiveness in capturing local details and tissue deformations. In this paper, we introduce UW-DNeRF, a novel approach utilizing neural radiance fields for high-quality reconstruction of deformable tissues. We propose an uncertainty-guided depth supervision strategy to mitigate the impact of inaccurate depth information. This strategy relaxes hard depth constraints and unlocks the potential of implicit neural representations. In addition, we design a local window-based information sharing scheme. This scheme employs local window and keyframe deformation networks to construct deformations with local awareness and enhances the model's ability to capture fine details. We demonstrate the superiority of our method over state-of-The-Art approaches on synthetic and in-vivo endoscopic datasets. Code is available at: https://***/IRMVLab/UW-DNeRF. © 2025 IEEE.

关键词： Endoscopy

来源：评论

学校读者我要写书评

暂无评论

Ultrahigh capacitive energy storage through dendritic nanopolar design

引用

Science (New York, N.Y.) 2025年第6743期388卷 211-216页

作者： Liu, Yajing Zhang, Yang Wang, Jing Yang, Chao Wang, Hongguang MacManus-Driscoll, Judith L. Yang, Hao van Aken, Peter A. Li, Weiwei Nan, Ce-Wen College of Physics MIIT Key Laboratory of Aerospace Information Materials and Physics State Key Laboratory of Mechanics and Control for Aerospace Structures Nanjing University of Aeronautics and Astronautics Nanjing China College of Integrated Circuits MIIT Key Laboratory of Aerospace Integrated Circuits and Microsystem Nanjing University of Aeronautics and Astronautics Nanjing China College of Aerospace Engineering State Key Laboratory of Mechanics and Control for Aerospace Structures Nanjing University of Aeronautics and Astronautics Nanjing China Max Planck Institute for Solid State Research Stuttgart Germany Department of Materials Science and Metallurgy University of Cambridge Cambridge United Kingdom State Key Laboratory of New Ceramics and Fine Processing School of Materials Science and Engineering Tsinghua University Beijing China

Electrostatic dielectric capacitors with ultrahigh power densities are sought after for advanced electronic and electrical systems owing to their ultrafast charge-discharge capability. However, low energy density resulting from low breakdown strength and suppressed polarization still remains a daunting challenge for practical applications. We propose a microstructural strategy with dendritic nanopolar (DNP) regions self-assembled into an insulator, which simultaneously enhances breakdown strength and high-field polarizability and minimizes energy loss and thus markedly improves energy storage performance and stability. For illustration, in this study, we achieved a high energy density of 215.8 joules per cubic centimeter with an efficiency of 80.7% at a high electric field of 7.4 megavolts per centimeter in a DNP structure-designed PbZr0.53Ti0.47O3-MgO film. The proposed strategy is generally applicable for development of high-performance dielectric microcapacitors.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dynamic and static Lagrange multiplier μ based Bayesian CP factorization with orthogonal factors: Theory and applications

引用

Signal processing 2025年 236卷

作者： Zhou, Jing Zhang, Zhichao School of Mathematics and Statistics Center for Applied Mathematics of Jiangsu Province and Jiangsu International Joint Laboratory on System Modeling and Data Analysis Nanjing University of Information Science and Technology Nanjing210044 China Key Laboratory of System Control and Information Processing Ministry of Education Shanghai Jiao Tong University Shanghai200240 China Key Laboratory of Computational Science and Application of Hainan Province Hainan Normal University Haikou571158 China

The existing μ-Singular Value Decomposition (μ-SVD) denoising algorithm is capable of extracting gear fault information under strong noise conditions. However, this algorithm is only applicable to two-dimensional real-valued data and lacks a mechanism for implementing Automatic Rank Determination (ARD) in high-dimensional data. In this paper, a Bayesian and Tensor treatment of μ-SVD is employed to enable ARD. To further investigate the impact of the Lagrange multiplier μ on the proposed μ-variational Bayesian (μ-VB) algorithm, we examine its performance from both static and dynamic perspectives. Simulation results demonstrate that the μ-VB algorithm achieves ARD and performs well in noise reduction. Further the μ-VB algorithm performs better in wireless communication and linear image coding across numerical domains, tensor sizes, orthogonal factors, and μ settings. © 2025 Elsevier B.V.

关键词： Tensors

来源：评论

学校读者我要写书评

暂无评论

Communication-Efficient Personalized Distributed Learning with Data and Node Heterogeneity

引用

IEEE Transactions on Cognitive Communications and Networking 2025年

作者： Tian, Zhuojun Zhang, Zhaoyang Li, Yiwei Bennis, Mehdi Zhejiang University College of Information Science and Electronic Engineering Hangzhou310027 China University of Oulu Center for Wireless Communications Oulu90014 Finland State Key Laboratory of Industrial Control Technology Hangzhou310027 China Zhejiang Provincial Key Laboratory of Multimodal Communication Networks and Intelligent Information Processing Hangzhou310027 China Xiamen University of Technology Fujian Key Laboratory of Communication Network and Information Processing Xiamen361024 China

To jointly tackle the challenges of data and node heterogeneity in decentralized learning, we propose a distributed strong lottery ticket hypothesis (DSLTH), based on which a communication-efficient personalized learning algorithm is developed. In the proposed method, each local model is represented as the ‌Hadamard product of global real-valued parameters and a personalized binary mask for pruning. The local model is learned by updating and fusing the personalized binary masks while the real-valued parameters are fixed among different agents. To further reduce the complexity of hardware implementation, we incorporate a group sparse regularization term in the loss function, enabling the learned local model to achieve structured sparsity. Then, a binary mask aggregation algorithm is designed by introducing an intermediate aggregation tensor and adding a personalized fine-tuning step in each iteration, which constrains model updates towards the local data distribution. The proposed method effectively leverages the relativity among agents while meeting personalized requirements in heterogeneous node conditions. We also provide a theoretical proof for the DSLTH, establishing it as the foundation of the proposed method. Numerical simulations confirm the validity of the DSLTH and demonstrate the effectiveness of the proposed algorithm. © 2015 IEEE.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

Event-based Video Person Re-identification via Cross-Modality and Temporal Collaboration

Event-based Video Person Re-identification via Cross-Modalit...

引用

International Conference on Acoustics, Speech, and Signal processing (ICASSP)

作者： Renkai Li Xin Yuan Wei Liu Xin Xu School of Computer Science and Technology Wuhan University of Science and Technology Hubei Province Key Laboratory of Intelligent Information Processing and Real-Time Industrial System

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Video-based person re-identification (ReID) has become increasingly important due to its applications in video surveillance applications. By employing events in video-based person ReID, more motion information can be provided between continuous frames to improve recognition accuracy. Previous approaches have assisted by introducing event data into the video person ReID task, but they still cannot avoid the privacy leakage problem caused by RGB images. In order to avoid privacy attacks and to take advantage of the benefits of event data, we consider using only event data. To make full use of the information in the event stream, we propose a Cross-Modality and Temporal Collaboration (CMTC) network for event-based video person ReID. First, we design an event transform network to obtain corresponding auxiliary information from the input of raw events. Additionally, we propose a differential modality collaboration module to balance the roles of events and auxiliaries to achieve complementary effects. Furthermore, we introduce a temporal collaboration module to exploit motion information and appearance cues. Experimental results demonstrate that our method outperforms others in the task of event-based video person ReID.

关键词： Data privacy Correlation Event detection Collaboration Transforms Streaming media Signal processing Video surveillance Speech processing Identification of persons

来源：评论

学校读者我要写书评

暂无评论

Fast resistivity imaging of transient electromagnetic using an extreme learning machine

引用

Soft Computing 2025年第2期29卷 1121-1131页

作者： Li, Ruiyou Zhang, Yong Li, Guang Li, Ruiheng Hu, Jia Li, Min School of Software and Internet of Things Engineering Jiangxi University of Finance and Economics NanChang330013 China School of Geophysics and Measurement Control Technology East China University of Technology NanChang330013 China School of Information Engineering Hubei University of Economics Wuhan430205 China The State Key Laboratory of Power Transmission Equipment and System Security and New Technology Chongqing University Chongqing400044 China

The transient electromagnetic method (TEM) is widely used in geophysical exploration. In TEM data interpretation, nonlinear inversion plays an important role. However, traditional TEM nonlinear inversion adopts the OCCAM imaging method, which merely presents the approximate shape of the stratum model, with poor inversion accuracy and much iteration time. To solve the above problems, a novel nonlinear inversion approach based on extreme learning machine (ELM) is proposed in this paper. The ELM is required to establish the input–output mapping relationship of the inversion network only once through the analytical method, which is different from the traditional neural network method that demands iterative gradient learning and is prone to fall into the local optimum. Moreover, the ELM inversion network by randomly assigning the hidden layer parameters is capable of mapping the observed TEM data and quickly producing resistivity images, which avoids time-consuming iteration and inversion calculations. The presented approach is applied to both synthetic and field examples. The results show that compared with the traditional nonlinear inversion algorithms (BP and OCCAM), the proposed method achieves better inversion accuracy and significantly reduces the calculation time, which verifies the effectiveness of the ELM algorithm for TEM data interpretation. Additionally, the research provides a new method and technology for TEM data inversion. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Electromagnetic transients

来源：评论

学校读者我要写书评

暂无评论

VAGeo: View-specific Attention for Cross-View Object Geo-Localization

VAGeo: View-specific Attention for Cross-View Object Geo-Loc...

引用

International Conference on Acoustics, Speech, and Signal processing (ICASSP)

作者： Zhongyang Li Xin Yuan Wei Liu Xin Xu School of Computer Science and Technology Wuhan University of Science and Technology Hubei Province Key Laboratory of Intelligent Information Processing and Real-Time Industrial System

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Cross-view object geo-localization (CVOGL) aims to locate an object of interest in a captured ground- or drone-view image within the satellite image. However, existing works treat ground-view and drone-view query images equivalently, overlooking their inherent viewpoint discrepancies and the spatial correlation between the query image and the satellite-view reference image. To this end, this paper proposes a novel View-specific Attention Geo-localization method (VAGeo) for accurate CVOGL. Specifically, VAGeo contains two key modules: view-specific positional encoding (VSPE) module and channel-spatial hybrid attention (CSHA) module. In object-level, according to the characteristics of different viewpoints of ground and drone query images, viewpoint-specific positional codings are designed to more accurately identify the click-point object of the query image in the VSPE module. In feature-level, a hybrid attention in the CSHA module is introduced by combining channel attention and spatial attention mechanisms simultaneously for learning discriminative features. Extensive experimental results demonstrate that the proposed VAGeo gains a significant performance improvement, i.e., improving acc@0.25/acc@0.5 on the CVOGL dataset from 45.43%/42.24% to 48.21%/45.22% for ground-view, and from 61.97%/57.66% to 66.19%/61.87% for drone-view.

关键词： Location awareness Accuracy Attention mechanisms Image coding Signal processing Encoding Satellite images Object recognition Speech processing Drones

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：