检索结果-内蒙古大学图书馆

On large language models safety,security,and privacy:A survey

Journal of Electronic science and technology 2025年第1期23卷 1-21页

作者： Ran Zhang Hong-Wei Li Xin-Yuan Qian Wen-Bo Jiang Han-Xiao Chen School of Computer Science and Engineering University of Electronic Science and Technology of ChinaChengdu611731China

The integration of artificial intelligence(AI)technology,particularly large language models(LLMs),has become essential across various sectors due to their advanced language comprehension and generation *** their transformative impact in fields such as machine translation and intelligent dialogue systems,LLMs face significant *** challenges include safety,security,and privacy concerns that undermine their trustworthiness and effectiveness,such as hallucinations,backdoor attacks,and privacy *** works often conflated safety issues with security *** contrast,our study provides clearer and more reasonable definitions for safety,security,and privacy within the context of *** on these definitions,we provide a comprehensive overview of the vulnerabilities and defense mechanisms related to safety,security,and privacy in ***,we explore the unique research challenges posed by LLMs and suggest potential avenues for future research,aiming to enhance the robustness and reliability of LLMs in the face of emerging threats.

关键词： Large language models Privacy issues Safety issues Security issues

来源：评论

学校读者我要写书评

暂无评论

Multi-Scale Feature Optimization Point Cloud Completion Network Integrating SoftPool

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer science 2025年第1期52卷 233-243页

作者： Zhang, Wanpeng Zhou, Ziwei School of Computer Science and Software Engineering University of Science and Technology LiaoNing Anshan114051 China School of Computer Science and Software Engineering University of Science and Technology LiaoNing Anshan114051 China

Point cloud completion is crucial in point cloud processing, as it can repair and refine incomplete 3D data, ensuring more accurate models. However, current point cloud completion methods commonly face a challenge: they fail to fully utilize multi-scale information from local features, leading to limitations in accuracy and detail preservation. To address this issue, this paper proposes a multi-scale feature optimization algorithm for point cloud completion that integrates SoftPool. Based on DGCNN, the method combines dilated convolution and bottleneck attention mechanisms to extract features at different scales, enhancing the ability to capture detailed information in point clouds. The bottleneck attention mechanism is used to optimize important detail features. The extracted local features are concatenated with their corresponding positional information to form point proxies, enhancing the effective extraction of local geometric features, resulting in more refined completed point cloud shapes. A Transformer architecture is employed to model these features. Finally, SoftPool is introduced for fine-grained feature downsampling, improving the network's ability to recover point cloud details. FoldingNet is used to reconstruct missing structures and output the completed point cloud. To validate the model's completion performance, training and testing are conducted on the PCN and ShapeNet55 datasets. Experimental results demonstrate that the model has better feature detail retention and more accurate completion results. On the PCN dataset, the average CD value is reduced by 6.5% compared to the best-performing model among the comparison methods. On the ShapeNet55 dataset, the average CD value across three difficulty levels is reduced by 6.9% compared to the best-performing model among the comparison methods. Additionally, the model also achieved a 2.1% improvement in F-score. © (2025), (International Association of Engineers). All rights reserved.

关键词： Ability testing

来源：评论

学校读者我要写书评

暂无评论

Effect of Nitrogen and Sulfur Co‑Doped Graphene on the Electrochemical Hydrogen Storage Performance of Co_(0.9)Cu_(0.1)Si Alloy

引用

Acta Metallurgica Sinica(English Letters) 2023年第6期36卷 1023-1037页

作者： Wenhao Fan Jianxun Zhao Dayong Liu Qingcheng Liang Wanqiang Liu Qingshuang Wang Heng Liu Peng Chen Shang Gao Xinlong Bao Yong Cheng Xinwei Wang Xin Guo School of Materials Science and Engineering Changchun University of Science and TechnologyChangchun 130022China Engineering Research Center of Optoelectronic Functional Materials Ministry of EducationChangchun 130022China College of Electronic Science and Engineering Jilin UniversityChangchun 130022China School of Life Science and Technology Changchun University of Science and TechnologyChangchun 130022China Research Center for Nanotechnology Changchun University of Science and TechnologyChangchun 130022China Research Institute of Changchun University of Science and Technology in Chongqing Chongqing 401120China State Key Laboratory of Rare Earth Resource Utilization Changchun Institute of Applied ChemistryCASChangchun 130022China

Co_(0.9)Cu_(0.1)Si alloy was prepared by mechanical alloying ***-doped graphene(NG)and nitrogen–sulfur codoped graphene(NSG)were prepared by hydrothermal method.5 wt%graphene oxide,NG and NSG were doped into Co_(0.9)Cu_(0.1)Si alloy,respectively,by ball milling to improve the electrochemical hydrogen storage performance of the composite material.X-ray diffraction and scanning electron microscopy were used to characterize the structure and morphology of the composite material,and the LAND battery test system and three-electrode battery system were used to test the electrochemical performance of the composite *** composite material showed better discharge capacity and better cycle stability than the pristine *** addition,in order to study the optimal ratio of NSG,3%,5%,7%and 10%of NSG were doped into Co_(0.9)Cu_(0.1)Si alloy,***_(0.9)Cu_(0.1)Si alloy doped with 5%NSG had the best performance among all the *** best discharge capacity was 580.1 mAh/g,and its highest capacity retention rate was 64.1%.The improvement in electrochemical hydrogen storage performance can be attributed to two *** the one hand,the electrocatalytic performance of graphene is improved by co-doping nitrogen and sulfur,on the other hand,graphene has excellent electrical conductivity.

关键词： Co_(0.9)Cu_(0.1)Si alloy Nitrogen-sulfur co-doped graphene(NSG) Composite material Electrochemical hydrogen storage

来源：评论

学校读者我要写书评

暂无评论

Improving UAV Image Target Detection: A Novel Approach Using OptiDETR with Swin Transformer

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer science 2025年第3期52卷 771-780页

作者： Ma, Wenlong Liu, Weisheng School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China

In the analysis of drone aerial images, object detection tasks are particularly challenging, especially in the presence of complex terrain structures, extreme differences in target sizes, suboptimal shooting angles, and varying lighting conditions, all of which exacerbate the difficulty of recognition. In recent years, the DETR model based on the Transformer architecture has eliminated traditional post-processing steps such as NMS(Non-Maximum Suppression), thereby simplifying the object detection process and improving detection accuracy, which has garnered widespread attention in the academic community. However, DETR has limitations such as slow training convergence, difficulty in query optimization, and high computational costs, which hinder its application in practical fields. To address these issues, this paper proposes a new object detection model called OptiDETR. This model first employs a more efficient hybrid encoder to replace the traditional Transformer encoder. The new encoder significantly enhances feature processing capabilities through internal and cross-scale feature interaction and fusion logic. Secondly, an IoU (Intersection over Union) aware query selection mechanism is introduced. This mechanism adds IoU constraints during the training phase to provide higher-quality initial object queries for the decoder, significantly improving the decoding performance. Additionally, the OptiDETR model integrates SW-Block into the DETR decoder, leveraging the advantages of Swin Transformer in global context modeling and feature representation to further enhance the performance and efficiency of object detection. To tackle the problem of small object detection, this study innovatively employs the SAHI algorithm for data augmentation. Through a series of experiments, It achieved a significant performance improvement of more than two percentage points in the mAP (mean Average Precision) metric compared to current mainstream object detection models. Furthermore, ther

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

OCRBench: on the hidden mystery of OCR in large multimodal models

引用

science China(Information sciences) 2024年第12期67卷 23-35页

作者： Yuliang LIU Zhang LI Mingxin HUANG Biao YANG Wenwen YU Chunyuan LI Xu-Cheng YIN Cheng-Lin LIU Lianwen JIN Xiang BAI School of Artificial Intelligence and Automation Huazhong University of Science and Technology School of Electronic and Information Engineering South China University of Technology Microsoft Research School of Computer & Communication Engineering University of Science and Technology Beijing Institute of Automation Chinese Academy of Sciences School of Software Engineering Huazhong University of Science and Technology

Large models have recently played a dominant role in natural language processing and multimodal vision-language learning. However, their effectiveness in text-related visual tasks remains relatively unexplored. In this paper, we conducted a comprehensive evaluation of large multimodal models, such as GPT4V and Gemini, in various text-related visual tasks including text recognition, scene text-centric visual question answering(VQA), document-oriented VQA, key information extraction(KIE), and handwritten mathematical expression recognition(HMER). To facilitate the assessment of optical character recognition(OCR) capabilities in large multimodal models, we propose OCRBench, a comprehensive evaluation benchmark. OCRBench contains 29 datasets, making it the most comprehensive OCR evaluation benchmark available. Furthermore, our study reveals both the strengths and weaknesses of these models, particularly in handling multilingual text, handwritten text, non-semantic text, and mathematical expression *** importantly, the baseline results presented in this study could provide a foundational framework for the conception and assessment of innovative strategies targeted at enhancing zero-shot multimodal *** evaluation pipeline and benchmark are available at https://***/Yuliang-Liu/Multimodal OCR.

关键词： large multimodal model OCR text recognition scene text-centric VQA document-oriented VQA key information extraction handwritten mathematical expression recognition

来源：评论

学校读者我要写书评

暂无评论

Design of modified model of intelligent assembly digital twins based on optical fiber sensor network

引用

Digital Communications and Networks 2024年第5期10卷 1542-1552页

作者： Zhichao Liu Jinhua Yang Juan Wang Lin Yue School of Opto-Electronic Engineering Changchun University of Science and TechnologyChangchun130000JilinChina Key Laboratory of Optoelectric Measurement and Optical Information Transmission Technology of Ministry of Education Changchun University of Science and TechnologyChangchun130000JilinChina Department of Actuarial Science and Business Analytics Macquarie UniversityNSW2109Australia

Intelligent assembly of large-scale,complex structures using an intelligent manufacturing platform represents the future development direction for industrial *** large-scale structural assembly processes,several bottleneck problems occur in the existing auxiliary assembly ***,the traditional LiDARbased assembly technology is often limited by the openness of the manufacturing environment,in which there are blind spots,and continuous online assembly adjustment thus cannot be ***,for assembly of large structures,a single-station LiDAR system cannot achieve complete coverage,which means that a multi-station combination method must be used to acquire the complete three-dimensional data;many more data errors are caused by the transfer between stations than by the measurement accuracy of a single station,which means that the overall system's measurement and adjustment errors are increased ***,because of the large numbers of structural components contained in a large assembly,the accumulated errors may lead to assembly interference,but the LiDAR-assisted assembly process does not have a feedback perception capability,and thus assembly component loss can easily be caused when assembly interference ***,this paper proposes to combine an optical fiber sensor network with digital twin technology,which will allow the test data from the assembly entity state in the real world to be applied to the"twin"model in the virtual world and thus solve the problems with test openness and data *** problem of station and perception feedback is also addressed and represents the main innovation of this *** system uses an optical fiber sensor network as a flexible sensing medium to monitor the strain field distribution within a complex area in real time,and then completes real-time parameter adjustment of the virtual assembly based on the distributed *** areas include areas that are laser-unreachable,areas wi

关键词： Digital twins Intelligent manufacturing Intelligent assembly Optical fiber sensor network Assembly condition monitoring algorithm

来源：评论

学校读者我要写书评

暂无评论

Adaptive coupled-sliding-variable-based finite-time control of composite formation for multi-robot systems

引用

science China(Information sciences) 2024年第10期67卷 244-260页

作者： Xinru MA Hengyu LI Jun LIU Yueying WANG Shaorong XIE Jun LUO School of Mechatronic Engineering and Automation Shanghai University School of Mathematics and Computer Application Technology Jining University School of Computer Science Shanghai University College of Mechanical and Vehicle Engineering Chongqing University

This paper focuses on the finite-time control(FTC) of the composite formation consensus(CFC)problems for multi-robot systems(MRSs). The CFC problems are firstly proposed for MRSs under the complex network topology of cooperative or cooperative-competitive networks. Regarding the problems of FTC and CFC on multiple Lagrange systems(MLSs), coupled sliding variables are introduced to deal with the robustness and consistent convergence. Then, the adaptive finite-time protocols are given based on the displacement approaches. With the premised FTC, tender-tracking methods are further developed for the problems of tracking information disparity. Stability analyses of those MLSs mentioned above are clarified with Lyapunov candidates considering the coupled sliding vectors, which provide new verification for tender-tracking systems. Under the given coupled-sliding-variable-based finite-time protocols, MLSs distributively adjust the local formation error to achieve global CFC and perform uniform convergence in time-varying tracking. Finally, simulation experiments are conducted while providing practical solutions for the theoretical results.

关键词： composite formation time-vary tracking information disparity tender-tracking finite-time control nonlinear robots complex networks symmetry consensus

来源：评论

学校读者我要写书评

暂无评论

Soft shadow rendering algorithm based on light characteristics of rectangle regions 4

Soft shadow rendering algorithm based on light characteristi...

引用

4th International Conference on computer Vision, Application, and Algorithm, CVAA 2024

作者： Yi, Huiping Chen, Chunyi Li, Hailan School of Computer Science and Technology Changchun University of Science and Technology Changchun130022 China

ISBN: (数字)9781510687622

ISBN: (纸本)9781510687615

Aiming at the difficulty that light soft shadow rendering in rectangular areas cannot meet the physical accuracy in real-time rendering, an approximate rectangular area light model rendering algorithm is proposed. The algorithm first generates shadow maps to record depth information from the perspective of the light source to render the scene, and then proposes a model that considers the relationship between the spatial position of the occlusion object and the shape of the light in the rectangular area. Calculate the width of the light in each direction of the rectangular area, the average depth around the texel, and the distance from the occluded object to the receiving object. Determine the spatial filter width of the shadow map corresponding to the visual field attractions, and finally use this width to obtain high-realism soft shadows in the shadow map. Experimental results show that this method is more effective than related algorithms. © 2025 SPIE.

关键词： Wiener filtering

来源：评论

学校读者我要写书评

暂无评论

Research on Influen e Mechanism of Nozzle Structure on Properties of Thick Carbon Steel Plate by Laser Cutting

引用

Chinese Journal of Mechanical engineering 2024年第6期37卷 118-130页

作者： Nan Chen Yan Shi Yan Pang Ruotao Li School of Mechanical and Electric Engineering Changchun University of Science and TechnologyChangchun 130022China Changchun College of Electronic Technology Changchun 130114China Hsg Laser Co. Ltd.Foshan 528306China

Aiming at problems such as slow cutting speed,poor cutting quality and inability to cut thick plates effectively with traditional single-layer nozzle structure,it is proposed to add an outer ring core to form a double-layer nozzle structure on the basis of the single-layer nozzle structure to improve its cutting *** three-dimensional entity of single-and double-layer nozzle is established,and the simulation model of auxiliary gas flow field is established by using N-S equation and SST k-ωturbulence model assisted by laser cutting in this *** simulation results show that under the same cutting parameters,the auxiliary gas flow rate,oxygen concentration,mass flow rate,shear force and dynamic pressure of the double-layer nozzle structure are higher than those of the single-layer nozzle *** addition,compared with the single-layer nozzle structure,the double-layer nozzle structure can effectively improve the attenuation of oxygen concentration in the direction of sheet *** nozzle structures were used to laser cutting carbon steel plates with a thickness of 20 mm in the experiment,and the experimental parameters of cutting speed were 1.2–1.6 m/*** comparison results show that the cutting speed and cutting section quality of the double-layer nozzle are better than those of the single-layer nozzle,and the results of simulation analysis are consistent with the cutting test results.

关键词： Laser cutting Nozzle structure Flow field analysis Cutting speed Section quality

来源：评论

学校读者我要写书评

暂无评论

Saliency detection method for Panoramic Images based on GCN-ELM 4

Saliency detection method for Panoramic Images based on GCN-...

引用

4th International Conference on computer Vision, Application, and Algorithm, CVAA 2024

作者： An, Nan Yu, Haiyang Zhang, Ripei Hu, Xiaojuan Li, Yanfeng School of Computer Science and Technology Changchun University of Science and Technology Changchun130022 China

ISBN: (数字)9781510687622

ISBN: (纸本)9781510687615

In the field of saliency detection for panoramic images, traditional equirectangular and cube projection methods in panoramic image saliency detection often face issues like distortion and discontinuities, impacting detection accuracy. This study introduces an innovative image resampling technique and a GCN-ELM joint model. By evenly distributing spherical pixels onto a 2D plane, the method reduces pixel redundancy from equirectangular projection. Experimental results show that this approach significantly enhances saliency detection performance compared to existing methods. © 2025 SPIE.

关键词： Adversarial machine learning

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：