检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2025年第2期6卷 405-415页

作者： Cai, Zinuo Ma, Rongbo Fu, Yicheng Zhang, Weishan Ma, Ruhui Guan, Haibing Shanghai Jiao Tong University School of Electronic Information and Electrical Engineering Shanghai200240 China Qingdao Institute of Software College of Computer Science and Technology Qingdao266555 China

In recent years, the emergence of large-language models (LLMs) has profoundly transformed our production and lifestyle. These models have shown tremendous potential in fields, such as natural language processing, speech recognition, and recommendation systems, and are increasingly playing crucial roles in applications such as human-computer interaction and intelligent customer service. Efficient inference solutions for LLMs in data centers have been extensively researched, with a focus on meeting users' quality of service requirements. In this article, we focus on two additional requirements that responsible LLM inference should meet under QoS conditions: security throughout the model execution process and low maintenance requirements for the inference system. Therefore, we propose LLMaaS, a trusted model inference platform based on a serverless computing platform aimed at providing inference as a service for LLMs. First, we design a trusted serverless computing platform based on software guard extension (SGX), which includes distributed identity verification and SGX device plugins to ensure the security and trustworthiness of the inference process. Additionally, to reduce the maintenance requirements of the system, we enhance the SGX-based deep learning computing framework, including replacing PyTorch and using a greedy algorithm for graph partitioning. We conduct tests on four typical large models, and the experimental results demonstrate that, with minimal overhead and user code modifications, we can ensure the security of model execution. © 2020 IEEE.

关键词： Quality of service

来源：评论

学校读者我要写书评

暂无评论

Enhancing Large Language Model Inference Efficiency via Lookahead Cache Filtering

Enhancing Large Language Model Inference Efficiency via Look...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Ou, Jie Chen, Yueming Jiang, Shuaihong Tian, Wenhong School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu China

ISBN: (纸本)9798350368741

The large Key-Value (KV) cache is a significant challenge in deploying Large Language Models (LLMs). Current research addressing these issues employs cache compression techniques, which we find suffer from information loss and the "lost-in-the-middle" problem. We propose the Lookahead Cache Filtering (LCF), which retains the full cache in host memory to keep complete details, using the Important Key-Value Lookahead Prediction by Approximate Sorting and the Gather-based Matrix Multiplication on CPU, to filter important KV cache to reduce the overhead of loading cache to GPU and improve the throughput of inference. Finally, we use the Multi-Scale Pyramid information Fusion mechanism to enhance information fusion to further improve the effectiveness of inference results. Experiments demonstrate that LCF effectively maintains accuracy without introducing additional latency while reducing memory requirements. Our code will be available on GitHub. © 2025 IEEE.

关键词： Approximate Sorting Important Key-Value Lookahead information Fusion Lost-in-the-Middle Operator Fusion on CPU

来源：评论

学校读者我要写书评

暂无评论

Chemical bond and dielectric response of Yb_(2)Si_(2)O_(7)ceramic with low dielectric constant at GHz and THz

引用

Journal of Advanced Ceramics 2025年第2期14卷 119-129页

作者： Yingxiang Li Zegui Hou Zixuan Fang Fei Wang Deyin Liang Zitao Shi Wei Liu Chuansheng He Bin Tang College of Information Engineering Shaoyang UniversityShaoyang 422000China School of Materials and Energy University of Electronic Science and Technology of ChinaChengdu 611731China School of Civil Engineering and Geomatics Southwest Petroleum UniversityChengdu 610500China State Key Laboratory of Electronic Thin Films and Integrated Devices University of Electronic Science and Technology of ChinaChengdu 611731China

The exploration of high-performance dielectric ceramics with dielectric constants lower than 10 is of great significance for the next generation of wireless *** this study,we reported the Yb_(2)Si_(2)O_(7)(YSO)dielectric ceramics synthesized using a facile solid-state reaction technique.X-ray diffraction(XRD)and transmission electron microscopy(TEM)characterization confirmed that YSO ceramics had a monoclinic *** a sintering temperature of 1500℃,the sintered sample reached a maximum relative density of 96%and demonstrated the best dielectric properties:dielectric constant(ε_(r))=7.57,Q×f=78,645 GHz(Q is quality factor,and f is the resonant frequency of 13.5 GHz),andtemperature coefficient of the resonant frequency(τ_(f))=−13.5 ppm/℃.Moreover,Raman analysis revealed that the Si–O bond dominated the lattice vibration,and the full width at half maximum(FWHM)of the strong peak at 922 cm^(−1)was inversely proportional to the Q×f *** to the Phillips–van Vechten–Levine(P–V–L)theory,the values of Q×f andτ_(f)are affected primarily by the Si–O bond,whileε_(r)is influenced mainly by the Yb–O *** ceramics demonstrated excellent dielectric properties in the terahertz band(ε_(r)=8.13,Q×f=112,758 GHz),making them promising candidates for future applications in the terahertz band.

关键词： Yb_(2)Si_(2)O_(7)(YSO) dielectric properties Phillips-van Vechten-Levine(P-V-L) terahertz

来源：评论

学校读者我要写书评

暂无评论

Exciton-polaritons in a 2D hybrid organic-inorganic perovskite microcavity with the presence of optical Stark effect

引用

Chinese Physics B 2024年第3期33卷 551-555页

作者： Kenneth Coker 郑楚媛 Joseph Roger Arhin Kwame Opuni-Boachie Obour Agyekum 张伟利 School of Information and Communication Engineering University of Electronic Science and Technology of ChinaChengdu 611731China Department of Electrical and Electronic Engineering Ho Technical UniversityHo 00233Ghana Telecommunication Engineering Department Kwame Nkrumah University of Science and TechnologyKumasi 00233Ghana

This study investigates the properties of exciton-polaritons in a two-dimensional(2D)hybrid organic-inorganic perovskite microcavity in the presence of optical Stark *** both steady and dynamic state analyses,strong coupling between excitons of perovskite and cavity photons is revealed,indicating the formation of polaritons in the perovskite ***,it is found that an external optical Stark pulse can induce energy shifts of excitons proportional to the pulse intensity,which modifies the dispersion characteristics of the polaritons.

关键词： exciton-polaritons perovskite microcavity optical Stark effect

来源：评论

学校读者我要写书评

暂无评论

Uncertainty-aware Correspondence Distillation for Deep Image Clustering

Uncertainty-aware Correspondence Distillation for Deep Image...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Chang, Luyao Chen, Leiting Zhou, Chuan School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu China

ISBN: (纸本)9798350368741

Image clustering is a challenging task in computer vision, with performance heavily dependent on the quality of feature representations due to the inherent complexity of images. However, current image clustering methods overlook the underlying semantic information during representation learning, leading to low-quality feature representations. Moreover, the absence of ground-truth labels amplifies the detrimental effects of unreliable data on semantic guidance, steering the model towards incorrect learning directions. In this work, we propose a novel deep image clustering method named Uncertainty-aware Correspondence Distillation (UCD) to address these issues. Specifically, we introduce the concept of representation correspondence to establish cross-level connections between instances and semantics, which is further employed as a distillation target to improve the network's feature learning by complementing semantic information. To mitigate unnecessary similarity penalties arising from unreliable data, we develop robust dynamic weights for semantic guidance by modeling the uncertainty of image semantics. Extensive experiments on five benchmark datasets demonstrate the superiority of the proposed method. The code is available at https://***/YL616/UCD. © 2025 IEEE.

关键词： Image clustering Knowledge distillation Representation learning Semantic guidance

来源：评论

学校读者我要写书评

暂无评论

Individualized Data Generation in Personalized Federated Learning

引用

IEEE Transactions on Mobile Computing 2025年第7期24卷 6628-6642页

作者： Cai, Yunyun Xi, Wei Shen, Yuhao Sun, Cerui Wang, Shuai Gong, Wei Zhao, Jizhong Xi'an Jiaotong University Faculty of Electronic and Information Engineering Xi'an China Southeast University School of Computer Science and Engineering Nanjing China University of Science and Technology of China School of Computer science and Technology Hefei China

Most Personalized Federated Learning (PFL) algorithms merge the model parameters of each client with other (similar or generic) model parameters to optimize the personalized model (PM). However, the merged model parameters in these algorithms may fit low relevance data, thereby limiting the performance of PM. In this paper, we generate similar data for each client through the collaboration of a generic model (GM) on the server, rather than merging model parameters. To train a generator capable of generating data for all classes on the server without real data, we employ the GM as the discriminator in adversarial training with the generator. Additionally, we introduce a similarity assessment metric, which allows for the assessment of the similarity between local data and data from other classes. Nevertheless, the presence of non-IID data among clients can weaken the performance of the GM, consequently impacting the training of the generator and similarity assessment. To address this issue, we design a directive mechanism so that GM can be optimized during adversarial training without the need for additional training. The experimental results validate the superiority of our algorithm over state-of-the-art algorithms in terms of accuracy, loss, and convergence speed. © 2002-2012 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Safe Tracking Control Strategy of Nonlinear Systems with Unknown Initial Tracking Condition: A Secure Boundary Protection Method Based on Prescribed Finite-time Control

引用

engineering Letters 2024年第7期32卷 1402-1411页

作者： Songnan, Zou Xiaohua, Li Yang, Liu School of Electronic and Information Engineering University of Science and Technology Liaoning TX Anshan114051 China College of Automation and Electronic Engineering Qingdao University of Science and Technology TX Qingdao266061 China

A safe tracking control problem is investigated for a class of nonlinear systems with unknown initial tracking condition. The safe tracking control is independent of the initial tracking condition with the help of a novel output mapping. A secure boundary protection method (SBPM) based on prescribed finite-time control and a new prescribed finite-time performance function (PFTPF) is proposed. The SBPM can guarantee the safe operation of the system when the desired output violates the output constraint function. The method can ensure the system satisfies both the tracking control performance specified by the PFTPF and the actual output constraints. It can effectively handle the abrupt changes of actual output constraints. To solve the excessive control input jitter problem caused by the sudden changes of the output con-straints, a bidirectional filtering smoothing mechanism (BFSM) is proposed. Finally, the effectiveness and superiority of the proposed method are verified by simulations. © 2024, International Association of Engineers. All rights reserved.

关键词： Nonlinear systems

来源：评论

学校读者我要写书评

暂无评论

Anti-swarm UAV radar system based on detection data fusion

引用

Journal of Systems engineering and electronics 2024年第5期35卷 1167-1176页

作者： WANG Pengfei HU Jinfeng HU Wen WANG Weiguang DONG Hao School of Information and Communication Engineering University of Electronic Science and Technology of ChinaChengdu 611731China China Airborne Missile Academy Luoyang 471009China Yangtze Delta Region Institute(Quzhou) University of Electronic Science and Technology of ChinaQuzhou 324003China College of Electronic and Information Engineering Nanjing University of Aeronautics and AstronauticsNanjing 211106China

There is a growing body of research on the swarm unmanned aerial vehicle(UAV)in recent years,which has the characteristics of small,low speed,and low height as radar *** confront the swarm UAV,the design of anti-UAV radar system based on multiple input multiple output(MIMO)is put forward,which can elevate the performance of resolution,angle accuracy,high data rate,and tracking flexibility for swarm UAV *** resolution and detection are the core problem in detecting the swarm *** distinct advantage of MIMO system in angular accuracy measurement is demonstrated by comparing MIMO radar with phased array *** MIMO radar has better performance in resolution,swarm UAV detection still has difficulty in target *** paper proposes a multi-mode data fusion algorithm based on deep neural networks to improve the detection ***,signal processing and data processing based on the detection fusion algorithm above are designed,forming a high resolution detection *** simulations are designed to illustrate the feasibility of the designed system and the proposed algorithm.

关键词： swarm radar high resolution deep neural network fusion algorithm

来源：评论

学校读者我要写书评

暂无评论

Multi-layer composite autoencoders for semi-supervised change detection in heterogeneous remote sensing images

引用

science China(information sciences) 2023年第4期66卷 124-125页

作者： Jiao SHI Tiancheng WU Hanwen YU A.K.QIN Gwanggil JEON Yu LEI School of Electronics and Information Northwestern Polytechnical University School of Resources and Environment University of Electronic Science and Technology of China Department of Computer Science and Software Engineering Swinburne University of Technology Department of Embedded Systems Engineering Incheon National University

With the increasing complexity of application scenarios, the fusion of different remote sensing data types has gradually become a trend, which can greatly improve the utilization of massive remote sensing *** the problem of change detection for heterogeneous remote images can be much more complicated than the traditional change detection for homologous remote sensing images,

关键词：

来源：评论

学校读者我要写书评

暂无评论

Distributed Hierarchical Multi-Agent Reinforcement Learning for Real-World Adversarial Team Games 21

Distributed Hierarchical Multi-Agent Reinforcement Learning ...

引用

21st International Computer Conference on Wavelet Active Media technology and information Processing, ICCWAMTIP 2024

作者： Wei, Li Bei, Hui Ke, Lu School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu611731 China School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu611731 China

ISBN: (纸本)9798331519254

Real-world applications often require teams of agents to coordinate strategies against adversaries. However, existing MARL models overlook resource constraints and task complexity, limiting their practical use. To address this, we propose the Distributed Hierarchical Multi-agent Learning (DHML) framework, which uses distributed neural networks for efficient computation. A mean-field mechanism and an attentional communication module are further incorporated for scalability, and a hierarchical task decomposition is adopted based on the option-critic framework for handling complexity. Experiments on a simulation platform modeling real communication networks demonstrate DHML's effectiveness across various adversarial scenarios. © 2024 IEEE.

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：