检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Xie, Qiujie Li, Qingqiu Yu, Zhuohao Zhang, Yuejie Zhang, Yue Yang, Linyi Zhejiang University China School of Engineering Westlake University China School of Computer Science Shanghai Key Lab of Intelligent Information Processing Shanghai Collaborative Innovation Center of Intelligent Visual Computing Fudan University China Peking University China Westlake Institute for Advanced Study China University College London United Kingdom Huawei Noah’s Ark Lab Hong Kong

As LLM-as-a-Judge emerges as a new paradigm for assessing large language models (LLMs), concerns have been raised regarding the alignment, bias, and stability of LLM evaluators. While substantial work has focused on alignment and bias, little research has concentrated on the stability of LLM evaluators. In this paper, we conduct extensive experiments involving 9 widely used LLM evaluators across 2 different evaluation settings to investigate the uncertainty in model-based LLM evaluations. We pinpoint that LLM evaluators exhibit varying uncertainty based on model families and sizes. With careful comparative analyses, we find that employing special prompting strategies, whether during inference or post-training, can alleviate evaluation uncertainty to some extent. By utilizing uncertainty to enhance LLM’s reliability and detection capability in Out-Of-Distribution (OOD) data, we further fine-tune an uncertainty-aware LLM evaluator named ConfiLM using a human-annotated fine-tuning set and assess ConfiLM’s OOD evaluation ability on a manually designed test set sourced from the 2024 Olympics. Experimental results demonstrate that incorporating uncertainty as additional information during the fine-tuning phase can largely improve the model’s evaluation performance in OOD scenarios. The code and data are released at: https://***/hasakiXie123/LLM-Evaluator-Uncertainty. © 2025, CC BY-NC-SA.

关键词： Digital elevation model

来源：评论

学校读者我要写书评

暂无评论

Compact Millimeter Wave Massive MIMO System Utilizing ESPAR

引用

IEEE Transactions on Communications 2025年

作者： Zhang, Chi Shen, Shanpu Li, Hongyu Ma, Dingfei Han, Zixiang Qian, Jun Clerckx, Bruno Murch, Ross Peng Cheng Laboratory Department of Broadband Communication Shenzhen518055 China University of Liverpool Department of Electrical Engineering and Electronics LiverpoolL69 3GJ United Kingdom Guangzhou511400 China Guangdong University of Technology College of Information Engineering Guangzhou China China Mobile Research Institute Future Research Lab Beijing100053 China The Hong Kong University of Science and Technology Department of Electronic and Computer Engineering Clear Water Bay Kowloon Hong Kong Imperial College London Department of Electrical and Electronic Engineering LondonSW7 2AZ United Kingdom The Hong Kong University of Science and Technology Department of Electronic and Computer Engineering Institute for Advanced Study Clear Water Bay Kowloon Hong Kong

In this work, we propose a compact millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) system utilizing electronically steerable parasitic array radiator (ESPAR). We analyze the system and channel models for the compact mmWave massive MIMO system using a beamspace formulation and demonstrate that we can optimize the spectral efficiency of the compact mmWave massive MIMO system by jointly adjusting the variable reactive loads in the ESPAR with a digital beamformer. We formulate the compact mmWave massive MIMO system optimization problem to maximize spectral efficiency and propose an unconstrained optimization based algorithm with an initialization method. The spectral and energy efficiencies of the compact mmWave massive MIMO system are evaluated in comparison to conventional mmWave massive MIMO systems of the same antenna size using full-digital, fully-connected hybrid, and partially-connected hybrid beamforming. The results show that the compact mmWave massive MIMO system provides higher spectral efficiency than partially-connected hybrid beamforming. On top of this, it can provide higher energy efficiencies of around 3.21, 2.88, and 1.22 times compared to full-digital, fully-connected hybrid, and partially-connected hybrid beamforming but with lower hardware complexity and lower cost. Therefore, compact mmWave massive MIMO systems are a promising and effective alternative to conventional massive MIMO systems in millimeter wave (mmWave) communications. © 1972-2012 IEEE.

关键词： Beamforming

来源：评论

学校读者我要写书评

暂无评论

3D-Printed Stainless Steel Electrodes for Advancing Mems Microbial Fuel Cells Toward Sustainable on-chip Energy

3D-Printed Stainless Steel Electrodes for Advancing Mems Mic...

引用

IEEE International Conference on Micro Electro Mechanical Systems

作者： Anwar Elhadad Guangfa Li Jiaqi Yang Dehao Liu Seokheun Choi Department of Electrical & Computer Engineering Bioelectronics & Microsystems Laboratory State University of New York at Binghamton Binghamton New York USA Department of Mechanical Engineering Integrated Computational Materials Design Lab State University of New York at Binghamton Binghamton New York USA Center for Research in Advanced Sensing Technologies and Environmental Sustainability State University of New York at Binghamton Binghamton New York USA

ISBN: (数字)9798331508890

ISBN: (纸本)9798331508906

This study introduces a stainless steel microbial anode fabricated via advanced laser powder bed fusion, marking a transformative step in microbial fuel cell (MFC) technology. By leveraging this cutting-edge metal 3D printing approach, we achieve a 3D microporous structure with superior electrical conductivity, outstanding corrosion resistance in aquatic environments, robust mechanical strength, and adaptability for surface modification—essential attributes for sustainable, high-performance MFCs. The assembled MFC demonstrates unparalleled power output, along with exceptional durability and stability over multiple cycles, outperforming existing MFCs based on alternative 3D-printed substrates. This advancement not only solidifies the role of MFCs as a sustainable energy solution for powering on-chip electronics but also opens new avenues for scalable, cost-effective bioelectronic systems that align with the demands of next-generation technology.

关键词： Powders Catalysts Fuel cells Laser stability Three-dimensional printing System-on-chip Steel Anodes Laser fusion Surface treatment

来源：评论

学校读者我要写书评

暂无评论

Development of yellow-light TiO2 integrated photonics

引用

Optics Letters 2025年第11期50卷 3489-3492页

作者： Zequn Chen Yiheng Tang Maoliang Wei Xiaojing Li Boshu Sun Yingchun Wu Sishuo Zou Ji Huang Ke Si Wei Gong Hongtao Lin Lan Li Institute of Advanced Technology Westlake Institute for Advanced Study Hangzhou 310024 China Zhejiang Key Laboratory of 3D Micro/Nano Fabrication and Characterization School of Engineering Westlake University Hangzhou 310030 China The State Key Lab of Brain-Machine Intelligence College of Information Science and Electronic Engineering Zhejiang University Hangzhou 310027 China MOE Frontier Science Center for Brain Science & Brain-Machine Integration Zhejiang University Hangzhou 310027 China Nanhu Brain-Computer Interface Institute Hangzhou 311100 China College of Optical Science and Engineering Zhejiang University Hangzhou 310027 China Westlake Institute for Optoelectronics Fuyang Hangzhou Zhejiang 311421 China

Visible-light photonic integrated circuits (PICs) are rapidly emerging as a crucial technology to overcome the scaling challenges in quantum information processing and biosensing. Titanium dioxide (TiO 2 ), with its high transparency in the visible range, is regarded as a promising material for facilitating the development of high-performance visible-light integrated systems. In this paper, we introduce a series of passive TiO 2 integrated devices, which include waveguides, multimode interferometers (MMIs), and thermo-optic switches (TOSs), monolithically integrated for the first time to our knowledge at a working wavelength of 589 nm. A propagation loss of 5.46 ± 0.17 dB/cm demonstrates a significant improvement over previous reports on yellow-light TiO 2 waveguides. This study marks the first fabrication of a TOS operating at 589 nm, featuring a TiO 2 core and polymer cladding. Notably, it achieves a power consumption (P π ) of 8.24 mW, which represents a threefold reduction compared to silicon nitride-based TOSs operating in the visible spectrum. These results offer new insight into the design and fabrication of visible waveguide-integrated devices, paving the way for future emerging applications, such as quantum information, sensors, and biomedical science.

关键词： Biosensors Effective refractive index Information processing Photonics Quantum information Titanium dioxide

来源：评论

学校读者我要写书评

暂无评论

MTMD: Multi-Scale Temporal Memory Learning and Efficient Debiasing Framework for Stock Trend Forecasting

引用

IEEE Transactions on Emerging Topics in Computational Intelligence 2025年第3期9卷 2151-2163页

作者： Wang, Mingjie Tian, Juanxi Zhang, Mingze Guo, Jianxiong Jia, Weijia BNU-HKBU United International College Guangdong Key Lab of Ai and Multi-Modal Data Processing Department of Computer Science Zhuhai519087 China Gonzaga University School of Engineering and Applied Sciences SpokaneWA99258 United States Beijing Normal University Advanced Institute of Natural Sciences Zhuhai519087 China BNU-HKBU United International College Guangdong Key Lab of Ai and Multi-Modal Data Processing Zhuhai519087 China

The endeavor of stock trend forecasting is principally focused on predicting the future trajectory of the stock market, utilizing either manual or technical methodologies to optimize profitability. Recent advancements in machine learning technologies have showcased their efficacy in discerning authentic profit signals within the realm of stock trend forecasting, predominantly employing temporal data derived from historical stock price patterns. Nevertheless, the inherently volatile and dynamic characteristics of the stock market render the learning and capture of multi-scale temporal dependencies and stable trading opportunities a formidable challenge. This predicament is primarily attributed to the difficulty in distinguishing real profit signal patterns amidst a plethora of mixed, noisy data. In response to these complexities, we propose a Multi-Scale Temporal Memory Learning and Efficient Debiasing (MTMD) model. This innovative approach encompasses the creation of a learnable embedding coupled with external attention, serving as a memory module through self-similarity. It aims to mitigate noise interference and bolster temporal consistency within the model. The MTMD model adeptly amalgamates comprehensive local data at each timestamp while concurrently focusing on salient historical patterns on a global scale. Furthermore, the incorporation of a graph network, tailored to assimilate global and local information, facilitates the adaptive fusion of heterogeneous multi-scale data. Rigorous ablation studies and experimental evaluations affirm that the MTMD model surpasses contemporary state-of-the-art methodologies by a substantial margin in benchmark datasets. © 2025 IEEE.

关键词： Profitability

来源：评论

学校读者我要写书评

暂无评论

Synthesized Acoustic Vortex-Frequency Comb via Rotational Doppler Effect

引用

Physical Review Letters 2025年第13期134卷 137001-137001页

作者： Yurou Jia Xuan Zhang Suying Zhang Guoqiang Xu Taoyu Chen Hongtao Zhou Yechao Bai Ying Cheng Dajian Wu Department of Physics MOE Key Laboratory of Modern Acoustics Collaborative Innovation Center of Advanced Microstructures Jiangsu Physical Science Research Center Nanjing University Nanjing 210093 China Department of Electrical and Computer Engineering National University of Singapore Singapore Department of Mechanics School of Mechanical Engineering Tianjin University Tianjin 300350 China School of Electronic Science and Engineering Nanjing University Nanjing 210023 China Jiangsu Key Lab on Opto-Electronic Technology School of Physics and Technology Nanjing Normal University Nanjing 210023 China

The synergistic control of light’s frequency and orbital angular momentum (OAM) via integrated nonlinear ring microresonators is crucial for creating spatiotemporal optical waveforms and advancing optical metrology. A direct analog of this capability in acoustics is challenging due to the nonlinearity-based mechanisms’ dependence on external driving and limited frequency conversion efficiency. Here, we present an acoustic synthesized comb structure derived from spatiotemporal metasurfaces, concurrently controlling frequency and OAM characteristics with high conversion efficiency and removing the constraints of critical driving frequency and power level in nonlinear mechanisms. By optimizing comb structures, energy distributions among OAM modes are adjusted to different frequency lines. We finally propose spatiotemporal metalens devices based on vortex-frequency combs, enabling synchronously multiple-order spatial differential operations for parallel data processing. This Letter is groundbreaking in integrating spectral- and spatial-domain acoustic waves, promising multifunctional imaging and advantages in spatiotemporal beams.

关键词： Vortex flow

来源：评论

学校读者我要写书评

暂无评论

On Cost Aware Cloudlet Placement for Mobile Edge Computing

引用

IEEE/CAA Journal of Automatica Sinica 2019年第4期6卷 926-937页

作者： Qiang Fan Nirwan Ansari IEEE Advanced Networking Lab Department of Electrical and Computer EngineeringNew Jersey Institute of TechnologyNewarkUSA

As accessing computing resources from the remote cloud inherently incurs high end-to-end(E2E)delay for mobile users,cloudlets,which are deployed at the edge of a network,can potentially mitigate this *** some research works focus on allocating workloads among cloudlets,the cloudlet placement aiming to minimize the deployment cost(i.e.,consisting of both the cloudlet cost and average E2E delay cost)has not been addressed effectively so *** locations and number of cloudlets have a crucial impact on both the cloudlet cost in the network and average E2E delay of ***,in this paper,we propose the Cost Aware cloudlet PlAcement in moBiLe Edge computing(CAPABLE)strategy,where both the cloudlet cost and average E2E delay are considered in the cloudlet *** solve this problem,a Lagrangian heuristic algorithm is developed to achieve the suboptimal *** cloudlets are placed in the network,we also design a workload allocation scheme to minimize the E2E delay between users and their cloudlets by considering the user *** performance of CAPABLE has been validated by extensive simulations.

关键词： Cloudlet placement mobile cloud computing mobile edge computing

来源：评论

学校读者我要写书评

暂无评论

ANALYTICAL TRANSIENT-RESPONSE OF CMOS INVERTERS

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-FUNDAMENTAL THEORY AND APPLICATIONS 1992年第1期39卷 42-44页

作者： KAYSSI, AI SAKALLAH, KA BURKS, TM Advanced Computer Architecture Lab Department of Electrical Engineering and Computer Science University of Michigan Ann Arbor MI USA

A general formula relating the waveform at the output of a CMOS inverter to the waveform at its input is derived. The formula is applied to three cases: a step input, a ramp input, and an exponential input. A one-dimensional functional dependence of the inverter propagation delay and output slew rate on circuit parameters is derived and an inverter macromodel for timing analysis is suggested and experimentally verified.

关键词： Transient analysis Transient response Inverters Semiconductor device modeling Circuits Capacitance Timing Space exploration Signal design MOS devices

来源：评论

学校读者我要写书评

暂无评论

Adaptive Backdoor Attacks with Reasonable Constraints on Graph Neural Networks

引用

IEEE Transactions on Dependable and Secure Computing 2025年

作者： Dong, Xuewen Li, Jiachen Li, Shujun You, Zhichao Qu, Qiang Kholodov, Yaroslav Shen, Yulong The School of Computer Science and Technology Xidian University The Engineering Research Center of Blockchain Technology Application and Evaluation Ministry of Education China The Shaanxi Key Laboratory of Blockchain and Secure Computing Xi’an710071 China The School of Computing Kent Interdisciplinary Research Centre in Cyber Security University of Kent CanterburyCT2 7NF United Kingdom Shenzhen Institute of Advanced Technology Chinese Academy of Sciences China The Intelligent Transportation Systems Lab Innopolis University Innopolis Russia The School of Computer Science and Technology Xidian University China

Recent studies show that graph neural networks (GNNs) are vulnerable to backdoor attacks. Existing backdoor attacks against GNNs use fixed-pattern triggers and lack reasonable trigger constraints, overlooking individual graph characteristics and rendering insufficient evasiveness. To tackle the above issues, we propose ABARC, the first Adaptive Backdoor Attack with Reasonable Constraints, applying to both graph-level and node-level tasks in GNNs. For graph-level tasks, we propose a subgraph backdoor attack independent of the graph’s topology. It dynamically selects trigger nodes for each target graph and modifies node features with constraints based on graph similarity, feature range, and feature type. For node-level tasks, our attack begins with an analysis of node features, followed by selecting and modifying trigger features, which are then constrained by node similarity, feature range, and feature type. Furthermore, an adaptive edge-pruning mechanism is designed to reduce the impact of neighbors on target nodes, ensuring a high attack success rate (ASR). Experimental results show that even with reasonable constraints for attack evasiveness, our attack achieves a high ASR while incurring a marginal clean accuracy drop (CAD). When combined with the state-of-the-art defense randomized smoothing (RS) method, our attack maintains an ASR over 94%, surpassing existing attacks by more than 7%. © 2004-2012 IEEE. All rights reserved.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Novel WDM-PON architecture for simultaneous transmission of unicast data and multicast services

引用

Chinese Optics Letters 2010年第10期8卷 972-975页

作者：朱敏肖石林郭薇胡卫生 Benoit Geller State Key Lab of Advanced Optical Communication Systems and Networks Department of Electronic Engineering Shanghai Jiao Tong University Lab of Electronics and Computer Engineering ENSTA ParisTech

We propose a novel wavelength-division multiplexed passive optical network （WDM-PON） to simultaneously transmit unicast data and multicast services with upstream data re-modulation in optical network units （ONUs）. For each wavelength channel in the optical line terminal （OLT）, the downstream unicast data are applied to one arm of a dual-parallel Mach-Zehnder modulator （DPMZM） to generate baseband unicast non-return-to-zero （NRZ） signal. A radio frequency （RF） control signal is applied to the other arm to present two un-modulated sidebands for multicast data modulation in a differential phase-shift keying （DPSK） format. The flexible and dynamic multicast services are realized by simply switching the RF control signal on or off. The proposed scheme is experimentally demonstrated with 1.25-Gb/s downstream unicast, multicast, and upstream data.

关键词： Fiber optic networks Light modulators Multicasting Wavelength division multiplexing

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：