检索结果-内蒙古大学图书馆

2024 Findings of the Association for Computational Linguistics, EMNLP 2024

作者： Yamaguchi, Atsuki Villavicencio, Aline Aletras, Nikolaos School of Computer Science University of Sheffield United Kingdom Department of Computer Science Institute of Data Science and Artificial Intelligence University of Exeter United Kingdom The Alan Turing Institute United Kingdom

ISBN: (纸本)9798891761681

The development of state-of-the-art generative large language models (LLMs) disproportionately relies on English-centric tokenizers, vocabulary and pre-training data. Despite the fact that some LLMs have multilingual capabilities, recent studies have shown that their inference efficiency deteriorates when generating text in languages other than English. This results in increased inference time and costs. Cross-lingual vocabulary adaptation (CVA) methods have been proposed for adapting models to a target language aiming to improve downstream performance. However, the effectiveness of these methods on increasing inference efficiency of generative LLMs has yet to be explored. In this paper, we perform an empirical study of five CVA methods on four generative LLMs (including monolingual and multilingual models) across four typologically-diverse languages and four natural language understanding tasks. We find that CVA substantially contributes to LLM inference speedups of up to 271.5%. We also show that adapting LLMs that have been pre-trained on more balanced multilingual data results in downstream performance comparable to the original models. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

A near-zero quiescent power breeze wake-up anemometer based on a rolling-bearing triboelectric nanogenerator

引用

Microsystems & Nanoengineering 2024年第2期10卷 241-249页

作者： Xianpeng Fu Zhichao Jiang Jie Cao Zefang Dong Guoxu Liu Meiling Zhu Chi Zhang CAS Center for Excellence in Nanoscience Beijing Key Laboratory of Micro-nano Energy and SensorBeijing Institute of Nanoenergy and NanosystemsChinese Academy of SciencesBeijing101400China School of Nanoscience and Engineering University of Chinese Academy of SciencesBeijing100049China Center on Nanoenergy Research Guangxi Colleges and Universities Key Laboratory of Blue Energy and Systems IntegrationSchool of Physical Science&TechnologyGuangxi UniversityNanning530004China Institute of Intelligent Flexible Mechatronics Jiangsu UniversityZhenjiang212013China College of Engineering Mathematics and Physical ScienceUniversity of ExeterExeterEX44QFUK

Wind sensors have always played an irreplaceable role in environmental information monitoring and are expected to operate with lower power consumption to extend service ***,we propose a breeze wake-up anemometer(B-WA)based on a rolling-bearing triboelectric nanogenerator(RB-TENG)with extremely low static *** B-WA consists of two RB-TENGs,a self-waking-up module(SWM),a signal processing module(SPM),and a wireless transmission *** two RB-TENGs are employed for system activation and wind-speed *** the ambient wind-speed exceeds 2 m/s,the wake TENG(W-TENG)and the SWM can wake up the system within 0.96 *** the same time,the SPM starts to calculate the signal frequency from the measured TENG(M-TENG)to monitor the wind speed with a sensitivity of 9.45 Hz/(m/s).After the wind stops,the SWM can switch off the B-WA within 0.52 s to decrease the system energy *** quiescent on-duty mode,the operating power of the B-WA is less than 30 nW,which can greatly extend the service lifetime of the *** integrating triboelectric devices and rolling bearings,this work has realized an ultralow quiescent power and self-waked-up wireless wind-speed monitoring system,which has foreseeable applications in remote weather monitoring,IoT nodes,and so on.

关键词： generator power service

来源：评论

学校读者我要写书评

暂无评论

Single Application Service Deployment in the Edge Environment Based on the E-CARGO Model 26

Single Application Service Deployment in the Edge Environmen...

引用

26th International Conference on computer Supported Cooperative Work in Design, CSCWD 2023

作者： Zhang, Senyue Xue, Ling Huang, Weiliang Zhao, Lu Li, Ke Shenyang Aerospace University School of Artificial Intelligence Shenyang110136 China Nanjing University of Posts and Telecommunications School of Computer Science Nanjing210023 China University of Exeter Department of Computer Science ExeterEX4 4QF United Kingdom

ISBN: (纸本)9798350331684

The popularization and application of 5G technology is about to open the era of global "data explosion". The traditional cloud computing model shows insufficient service support for resource-sensitive applications, especially in terms of latency, and edge computing can solve this problem by providing service support close to the user request side. This article formalizes the single application service deployment problem (SASDP) using the E-CARGO (Environment-Class, Agent, Role, Group, and Object) model. Through group role assignment (GRA), a high-satisfaction service deployment scheme for single application service deployment is designed, and satisfaction evaluation is established through delay to provide providers with satisfactory deployment solutions and achieve economic benefits. Finally, large-scale simulation experiments are carried out based on Python PuLP platform, and experiments show that our method is better than the baseline method in terms of overall satisfaction, user coverage and economy. © 2023 IEEE.

关键词： Edge computing

来源：评论

学校读者我要写书评

暂无评论

A Novel Approach to Recognising Physical Activity Intensity in Adults: Gravity-Based Acceleration Imaging with Vision Transformer

A Novel Approach to Recognising Physical Activity Intensity ...

引用

IEEE International Symposium on Biomedical Imaging

作者： Lin Wang Tianle Zhang Miaomin Wang Gang Wang Guohao Chen Zhuhui Li Faculty of Health and Life Sciences University of Exeter Exeter UK Computer Sciences University of Liverpool Liverpool UK Faculty of Environment Science and Economy University of Exeter Exeter UK School of Computer Science and Engineering Northeastern University Shenyang China

ISBN: (数字)9798331520526

ISBN: (纸本)9798331520533

Research on physical activity intensity (PAI) recognition often relies on tabular data, limiting the capture of global patterns compared to image-based data, while traditional models struggle with long-sequence dependencies. This study introduces a novel method that uses a gravity-based acceleration algorithm to generate images of different intensity levels, followed by classification with a Vision Transformer (ViT) model, aiming to enhance model performance and experimental efficiency. The method uses the Euclidean Norm Minus One to calculate gravity-based acceleration per second. Gramian Angular Field images are generated based on adult intensity thresholds and are subsequently classified using the ViT model. The results showed the ViT model's high performance in recognising different PAIs. Specifically, it demonstrated that using gravity-based images for recognising PAI improves accuracy compared to METs-based images (etabolic equivalents), achieving an overall accuracy of 98.77% compared to 97.52%. Additionally, this novel approach maintains high accuracy across different time windows, with accuracy ranging from 98.59% to 99.05%, further indicating the model's stability and robust performance across varying temporal windows.

关键词： computer vision Accuracy Image recognition Limiting Transformers Stability analysis Robustness Distance measurement Pattern recognition Monitoring

来源：评论

学校读者我要写书评

暂无评论

Clinical Insight-Augmented Multi-View Learning for Alzheimer's Detection in Retinal OCTA Images

Clinical Insight-Augmented Multi-View Learning for Alzheimer...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Zhang, Yuandi Hao, Jinkui Zheng, Botian Liu, Yonghuai Meng, Yanda Zhang, Jiong Chen, Yang Yang, Chunfeng Zhao, Yitian Ministry of Education Southeast University Nanjing China Ningbo Institute of Materials Technology and Engineering Chinese Academy of Sciences Laboratory of Advanced Theranostic Materials and Technology Ningbo China Edge Hill University Department of Computer Science Ormskirk United Kingdom University of Exeter Department of Computer Science Exeter United Kingdom

ISBN: (纸本)9798350386226

Alzheimer's disease (AD) poses a significant global challenge, with a notable absence of accessible and cost-effective diagnostic tools for widespread AD detection. The retina, mirroring the brain in anatomy and physiology, has emerged as a potential avenue for rapid AD identification through retinal imaging. The current retinal image-based AD detection methods usually focus primarily on the macular area, but ignore the potential value that the optic disc region may have for the detection task. In this study, we leverage both macular- and disc-centered OCTA images and propose a multi-region fusion framework for AD detection. Based on clinical evidence, we integrate handcrafted features into the framework to improve model performance and interpretability. Specifically, vascular morphological parameters extracted from the macular and disc regions are used as input to a revalued KNN model to improve predictive capabilities. Furthermore, recognizing the significance of extracting and utilizing complementary information from the macular and optic disc regions, we propose an uncertainty-guided strategy based on Dempster-Shefer Theory (DST) to fuse knowledge from different regions. This approach considers each region's forecast quality and significantly improves the effectiveness and robustness of the model. Through comparative analysis with existing methods, we have demonstrated that our method outperforms the state-of-the-art ones and provides more valuable pathological evidence for the association between retinal vascular changes and AD. © 2024 IEEE.

关键词： Neurodegenerative diseases

来源：评论

学校读者我要写书评

暂无评论

Direct preference-based evolutionary multi-objective optimization with dueling bandits 24

Direct preference-based evolutionary multi-objective optimiz...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Tian Huang Shengbo Wang Ke Li School of Computer Science and Engineering University of Electronic Science and Technology of China Department of Computer Science University of Exeter

ISBN: (纸本)9798331314385

The ultimate goal of multi-objective optimization (MO) is to assist human decisionmakers (DMs) in identifying solutions of interest (SOI) that optimally reconcile multiple objectives according to their preferences. Preference-based evolutionary MO (PBEMO) has emerged as a promising framework that progressively approximates SOI by involving human in the optimization-cum-decision-making process. Yet, current PBEMO approaches are prone to be inefficient and misaligned with the DM's true aspirations, especially when inadvertently exploiting mis-calibrated reward models. This is further exacerbated when considering the stochastic nature of human feedback. This paper proposes a novel framework that navigates MO to SOI by directly leveraging human feedback without being restricted by a predefined reward model nor cumbersome model selection. Specifically, we developed a clustering-based stochastic dueling bandits algorithm that strategically scales well to high-dimensional dueling bandits. The learned preferences are then transformed into a unified probabilistic format that can be readily adapted to prevalent EMO algorithms. This also leads to a principled termination criterion that strategically manages human cognitive loads and computational budget. Experiments on 48 benchmark test problems, including the RNA inverse design and protein structure prediction, fully demonstrate the effectiveness of our proposed approach.

关键词：

来源：评论

学校读者我要写书评

暂无评论

The implicit bias of gradient descent toward collaboration between layers: a dynamic analysis of multilayer perceptions 24

The implicit bias of gradient descent toward collaboration b...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Zheng Wang Geyong Min Wenjie Ruan Department of Computer Science University of Exeter School of Computer Science USTC

ISBN: (纸本)9798331314385

The implicit bias of gradient descent has long been considered the primary mechanism explaining the superior generalization of over-parameterized neural networks without overfitting, even when the training error is zero. However, the implicit bias toward adversarial robustness has rarely been considered in the research community, although it is crucial for the trustworthiness of machine learning models. To fill this gap, in this paper, we explore whether layers in neural networks collaborate to strengthen adversarial robustness during gradient descent. By quantifying this collaboration between layers using our proposed concept, co-correlation, we demonstrate a monotonically increasing trend in co-correlation, which implies a decreasing trend in adversarial robustness during gradient descent. Additionally, we observe different behaviours between narrow and wide neural networks during gradient descent. We conducted extensive experiments that verified our proposed theorems.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-Authority Ciphertext-Policy Attribute-based Encryption with Hidden Policy for Securing Internet-of-Vehicles

Multi-Authority Ciphertext-Policy Attribute-based Encryption...

引用

IEEE International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom)

作者： Jie Cui Jing Zhang Lu Wei Minghui Zhu Hong Zhong Geyong Min School of Computer Science and Technology Anhui University Hefei China Department of Computer Science University of Exeter Exeter UK

ISBN: (数字)9798331506209

ISBN: (纸本)9798331506216

With the rapid development of the Internet-of-Vehicles (IoV) technologies, security and privacy issues associated with IoV data sharing have become increasingly prominent. Although attribute-based encryption (ABE) schemes offer effective solutions to these problems, the prevalent single-point failure vulnerabilities and risks of user privacy leakage in existing ABE schemes must be addressed. To this end, an original hidden policy scheme based on multi-authority ciphertext policy ABE is proposed. This scheme validates users by introducing multiple attribute authorities and generating intermediate attribute keys, effectively dispersing single-point performance pressure. Simultaneously, partial policy-hiding techniques are developed to ensure efficient system operation while protecting user privacy. Furthermore, this scheme introduces a central authority to track potentially malicious attribute authorities, preventing them from continuously generating incorrect attribute intermediate keys, thereby maintaining the overall security of the system. Additionally, by updating and revoking attribute versions, a flexible attribute revocation mechanism is achieved to further enhance system flexibility. Through in-depth security and performance analyses, the scheme is proven to be both secure and efficient for securing IoV.

关键词： Privacy Data privacy Systems operation Encryption Performance analysis Security Internet of Vehicles

来源：评论

学校读者我要写书评

暂无评论

ON ACCELERATING DIFFUSION-BASED SAMPLING PROCESSES VIA IMPROVED INTEGRATION APPROXIMATION 12

ON ACCELERATING DIFFUSION-BASED SAMPLING PROCESSES VIA IMPRO...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Zhang, Guoqiang Niwa, Kenta Kleijn, W. Bastiaan Dept. of Computer Science University of Exeter United Kingdom Communication Science Labs NTT Japan School of ECS Victoria Univ. of Wellington New Zealand

A popular approach to sample a diffusion-based generative model is to solve an ordinary differential equation (ODE). In existing samplers, the coefficients of the ODE solvers are pre-determined by the ODE formulation, the reverse discrete timesteps, and the employed ODE methods. In this paper, we consider accelerating several popular ODE-based sampling processes (including DDIM, DPM-Solver++, and EDM) by optimizing certain coefficients via improved integration approximation (IIA). We propose to minimize, for each time step, a mean squared error (MSE) function with respect to the selected coefficients. The MSE is constructed by applying the original ODE solver for a set of fine-grained timesteps, which in principle provides a more accurate integration approximation in predicting the next diffusion state. The proposed IIA technique does not require any change of a pre-trained model, and only introduces a very small computational overhead for solving a number of quadratic optimization problems. Extensive experiments show that considerably better FID scores can be achieved by using IIA-DDIM, IIA-DPM-Solver++, and IIA-EDM than the original counterparts when the neural function evaluation (NFE) is small (i.e., less than 25). © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Integration

来源：评论

学校读者我要写书评

暂无评论

DF-EGM: Personalised Knowledge Tracing with Dynamic Forgetting and Enhanced Gain Mechanisms

DF-EGM: Personalised Knowledge Tracing with Dynamic Forgetti...

引用

IEEE International Conference on Networking, Sensing and Control

作者： Hongyun Wang Leilei Shi ZiXuan Han Lu Liu Xiang Sun Furqan Aziz School of Computer Science and communication Engineering Jiangsu University China Department of Computer Science University of Exeter Exeter UK School of Computing and Mathematical Sciences University of Leicester UK

ISBN: (数字)9798350365221

ISBN: (纸本)9798350365238

In current online education, Knowledge Tracing (KT) technology plays a pivotal role, in monitoring and updating students' evolving knowledge states throughout their educational journey. Previous knowledge tracing models like DKT and DKVMN, have made some progress in personalised learning. However, they fall short in fully capturing the dynamic changes in the learning process, including forgetting behaviour and individual learning gains. To address these issues, this paper introduces a novel knowledge tracing model named dynamic forgetting and enhanced gain mechanisms (DF-EGM), which builds on Deep-Irt. The model initially utilises a CART decision tree classifier to extract pre-classification labels by analyzing response times and learning abilities. Subsequently, the model incorporates forgetting factors and personalised learning gains. It introduces attempt counts and time intervals to simulate knowledge forgetting. and the realization of personalised learning gains is achieved through a deeply refined knowledge gain module. This module models the perceived difficulty of exercises by students and dynamically adjusts learning gains based on their answer performance and current knowledge state. Experiments on public datasets indicate that the DF-EGM model can more accurately track students' knowledge states.

关键词： Knowledge engineering Analytical models Education Sensors Time factors Decision trees Monitoring

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：