检索结果-内蒙古大学图书馆

6th International Conference on Communications, Information System and computer Engineering, CISCE 2024

作者： Yu, Liang Lai, Yangbing Mo, Jiayu Zeng, Yuze Tan, Guoshen Xu, Haonan College of Computer Guangdong University of Science and Technology Dongguan China AIoT Edge Computing Engineering Technology Research Center of Dongguan City Guangdong AIoT Application Innovation Joint Laboratory Dongguan China

ISBN: (纸本)9798350353174

The current standard practice for building fire safety relies heavily on regular lighting, emergency lighting, and smoke sensors. The management and control centers for these systems are typically separate and independent, resulting in a bulky and expensive local control system that is difficult to use and manage. Additionally, the traditional fixed instructions in fire protection systems do not allow for adaptive planning of optimal escape routes based on the location of the fire, leading to delays in evacuation and ultimately causing significant casualties. This paper presents the design of a smart fire warning and evacuation system based on edge computing utilizing LED dynamic indication devices. The system integrates LED dot-matrix technology to create a smart indicator and emergency light with dynamic adaptive indication capabilities. Additionally, it includes temperature, humidity, and smoke sensors to monitor the building's environmental conditions. Through Mesh wireless networking, time-series sensing data is aggregated to the edge gateway, where a miniature machine learning model trained by the Edge Impluse (EI) platform performs edge computing for predicting and reasoning about fire situations. The data is then uploaded to a cloud server and forwarded to mobile users via Cat.1 communication protocol for real-time monitoring. The system also incorporates an improved ant colony algorithm for automatic execution in the cloud and allows manual commands from mobile terminals to control adaptive and dynamic indication of indicator lights and emergency lights, guiding building personnel to the safest evacuation path. System testing demonstrates a high accuracy rate of 95% for fire prediction and inference, with an average inference time of less than 1 ms. This system provides a novel approach to adaptive building fire prediction and evacuation, enhancing safety measures for orderly and secure evacuations. © 2024 IEEE.

关键词： Ant colony optimization

来源：评论

学校读者我要写书评

暂无评论

OneBit: Towards Extremely Low-bit Large Language Models 38

OneBit: Towards Extremely Low-bit Large Language Models

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Xu, Yuzhuang Han, Xu Yang, Zonghan Wang, Shuo Zhu, Qingfu Liu, Zhiyuan Liu, Weidong Che, Wanxiang Research Center for Social Computing and Information Retrieval Harbin Institute of Technology Harbin China Department of Computer Science & Technology Tsinghua University Beijing China

Model quantification uses low bit-width values to represent the weight matrices of existing models to be quantized, which is a promising approach to reduce both storage and computational overheads of deploying highly anticipated LLMs. However, current quantization methods suffer severe performance degradation when the bit-width is extremely reduced, and thus focus on utilizing 4-bit or 8-bit values to quantize models. This paper boldly quantizes the weight matrices of LLMs to 1-bit, paving the way for the extremely low bit-width deployment of LLMs. For this target, we introduce a 1-bit model compressing framework named OneBit, including a novel 1-bit parameter representation method to better quantize LLMs as well as an effective parameter initialization method based on matrix decomposition to improve the convergence speed of the quantization framework. Sufficient experimental results indicate that OneBit achieves good performance (at least 81% of the non-quantized performance on LLaMA models) with robust training processes when only using 1-bit weight matrices. Code and checkpoints are available at https://***/xuyuzhuang11/OneBit. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Relating microstructure to magnetocaloric properties in RE_(36)Tb_(20)Co_(20)Al_(24)(RE=Gd,Dy or Ho)high-entropy metallic-glass microwires designed by binary eutectic clusters method

引用

Journal of Materials Science & technology 2023年第18期149卷 167-176页

作者： Hangboce Yin Jun-Qiang Wang Yongjiang Huang Hongxian Shen Shu Guo Hongbo Fan Juntao Huo Jianfei Sun School of Materials Science and Engineering Harbin Institute of TechnologyHarbin 150001China CAS Key Laboratory of Magnetic Materials and Devices and Zhejiang Province Key Laboratory of Magnetic Materials and Application TechnologyNingbo Institute of Materials Technology&EngineeringChinese Academy of SciencesNingbo 315201China Center for Analysis Measurement and Computing Harbin Institute of TechnologyHarbin 150001China Space Environment Simulation Research Infrastructure Harbin Institute of TechnologyHarbin 150001China

The new high-entropy metallic-glasses(HE-MGs)are designed by using Dy and Ho to replace Gd in Gd_(36)Tb_(20)Co_(20)Al_(24)alloy based on the binary eutectic clusters *** with the equiatomic Gd 25 Tb 25 Co 25 Al 25 HE-MG,the non-equiatomic RE_(36)Tb_(20)Co_(20)Al_(24)(RE=Gd,Dy,or Ho)alloys show bet-ter glass-forming ability,which is attributed to the deep binary eutectic compositions used for alloy *** RE_(36)Tb_(20)Co_(20)Al_(24)alloys undergo second-order magnetic *** extreme peak value of magnetic entropy change is obtained as 10.3 J kg^(-1) K-1(5 T)for the Ho_(36)Tb_(20)Co_(20)Al_(24)***-situ high-energy synchrotron X-ray diffraction was conducted to observe the microstructural difference among non-equiatomic samples at cryogenic *** results indicate that Gd_(36)Tb_(20)Co_(20)Al_(24)alloy possesses a relatively large average value of the dispersion of local clusters at a low-temperature ***,com-bined with the critical exponentβclose to 0.5 of Gd_(36)Tb_(20)Co_(20)Al_(24)alloy,leads to its widest working temperature span among non-equiatomic *** work successfully establishes the connection be-tween microstructure and magnetocaloric properties of HE-MGs,which is beneficial for understanding the physical mechanism of the magnetocaloric behaviors of HE-MGs.

关键词： High-entropy metallic-glass Magnetocaloric effect High energy synchrotron X-ray diffraction Cryogenic temperature

来源：评论

学校读者我要写书评

暂无评论

PSST: A Benchmark for Evaluation-driven Text Public-Speaking Style Transfer

PSST: A Benchmark for Evaluation-driven Text Public-Speaking...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Sun, Huashan Wu, Yixiao Ye, Yuhao Yang, Yizhe Li, Yinghao Li, Jiawei Gao, Yang School of Computer Science and Technology Beijing Institute of Technology Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Applications China

ISBN: (纸本)9798891761681

Language style is necessary for AI systems to understand and generate diverse human language ***, previous text style transfer primarily focused on sentence-level data-driven approaches, limiting exploration of potential problems in large language models (LLMs) and the ability to meet complex application *** overcome these limitations, we introduce a novel task called Public-Speaking Style Transfer (PSST), which aims to simulate humans to transform passage-level, official texts into a public-speaking *** in the analysis of real-world data from a linguistic perspective, we decompose public-speaking style into key sub-styles to pose challenges and quantify the style modeling capability of *** such intricate text style transfer, we further propose a fine-grained evaluation framework to analyze the characteristics and identify the problems of stylized *** experiments suggest that current LLMs struggle to generate public speaking texts that align with human preferences, primarily due to excessive stylization and loss of semantic information. © 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

Contextual Modeling for Document-level ASR Error Correction 30

Contextual Modeling for Document-level ASR Error Correction

引用

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Jiang, Jin Yin, Xunjian Wan, Xiaojun Peng, Wei Li, Rongjun Yang, Jingyuan Zhou, Yanquan Wangxuan Institute of Computer Technology Peking University China Beijing University of Posts and Telecommunications China Artificial Intelligence Application Research Center Huawei Technologies China

ISBN: (纸本)9782493814104

Contextual information, including the sentences in the same document and in other documents of the dataset, plays a crucial role in improving the accuracy of document-level ASR Error Correction (AEC), while most previous works ignore this. In this paper, we propose a context-aware method that utilizes a k-Nearest Neighbors (kNN) approach to enhance the AEC model by retrieving a datastore containing contextual information. We conduct experiments on two English and two Chinese datasets, and the results demonstrate that our proposed model can effectively utilize contextual information to improve document-level AEC. Furthermore, the context information from the whole dataset provides even better results. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Error correction

来源：评论

学校读者我要写书评

暂无评论

FedIERF: Federated Incremental Extremely Random Forest for Wearable Health Monitoring

引用

Journal of computer Science & technology 2023年第5期38卷 970-984页

作者：胡春雨忽丽莎袁林陆佃杰吕蕾陈益强 Key Laboratory of Computing Power Network and Information Security Ministry of EducationShandong Computer Science Center(National Supercomputer Center in Jinan)Qilu University of Technology(Shandong Academy of Sciences)Jinan 250353China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer ScienceJinan 250000China Shandong Provincial Key Laboratory for Novel Distributed Computer Software Technology Jinan 25000China Institute of Information Technology Hebei University of Economics and BusinessShijiazhuang 050061China School of Information Science and Engineering Shandong Normal UniversityJinan 250358China Institute of Computing Technology Chinese Academy of SciencesBeijing 100190China

Wearable health monitoring is a crucial technical tool that offers early warning for chronic diseases due to its superior portability and low power ***,most wearable health data is distributed across dfferent organizations,such as hospitals,research institutes,and companies,and can only be accessed by the owners of the data in compliance with data privacy *** first challenge addressed in this paper is communicating in a privacy-preserving manner among different *** second technical challenge is handling the dynamic expansion of the federation without model *** address the first challenge,we propose a horizontal federated learning method called Federated Extremely Random Forest(FedERF).Its contribution-based splitting score computing mechanism significantly mitigates the impact of privacy protection constraints on model *** on FedERF,we present a federated incremental learning method called Federated Incremental Extremely Random Forest(FedIERF)to address the second technical *** introduces a hardness-driven weighting mechanism and an importance-based updating scheme to update the existing federated model *** experiments show that FedERF achieves comparable performance with non-federated methods,and FedIERF effectively addresses the dynamic expansion of the *** opens up opportunities for cooperation between different organizations in wearable health monitoring.

关键词： federated learning incremental learning random forest wearable health monitoring

来源：评论

学校读者我要写书评

暂无评论

research on neutron source accelerators visual-real fusion fault diagnosis with Digital Twin 4

Research on neutron source accelerators visual-real fusion f...

引用

4th International Conference on Electronic Information Engineering and computer Communication, EIECC 2024

作者： Yang, Sheng Zhang, Xiaodan Liang, Lizhen Liu, Shaoqing Wang, Xiaolong Qinghai University School of Computer Technology and Applications Xining China Intelligent Computing and Application Laboratory of Qinghai Province Xining China Institute of Energy Hefei Comprehensive National Science Center Hefei China

ISBN: (纸本)9798331534622

To address the intelligent upgrading needs of neutron source accelerators, a digital twin visualization system was designed and developed based on digital twin technology. The virtual mapping model was constructed in the digital space using actual parameters of physical space equipment, establishing a mapping connection between the physical and digital spaces centered around twin data. And a multi-channel integrated learning loop diagnostic model with virtual-physical fusion was proposed. Focusing on the experimental operational data of the equipment, the diagnostic model features data monitoring and multi-channel diagnostic functions. Input thresholds for each diagnostic channel of the model were configured based on the data scale. The final diagnostic result with higher accuracy was obtained by comprehensively comparing the results of each channel, significantly improving the real-time performance and efficiency of the diagnostic process. The effectiveness of the multichannel diagnostic model was experimentally validated using vibration signals generated during equipment operation. Additionally, the digital twin visualization system implemented data information visualization and intelligent management of accelerator equipment. © 2024 IEEE.

关键词： Neutrons

来源：评论

学校读者我要写书评

暂无评论

A Sharding Scheme Based on Graph Partitioning Algorithm for Public Blockchain

引用

computer Modeling in Engineering & Sciences 2024年第6期139卷 3311-3327页

作者： Shujiang Xu Ziye Wang Lianhai Wang Miodrag J.Mihaljevi′c Shuhui Zhang Wei Shao Qizheng Wang Key Laboratory of Computing Power Network and Information Security Ministry of EducationShandong Computer Science CenterQilu University of Technology(Shandong Academy of Sciences)Jinan250014China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer ScienceJinan250014China Mathematical Institute The Serbian Academy of Sciences and ArtsBelgrade11000Serbia

Blockchain technology,with its attributes of decentralization,immutability,and traceability,has emerged as a powerful catalyst for enhancing traditional industries in terms of optimizing business ***,transaction performance and scalability has become the main challenges hindering the widespread adoption of *** to its inability to meet the demands of high-frequency trading,blockchain cannot be adopted in many *** improve the transaction capacity,researchers have proposed some on-chain scaling technologies,including lightning networks,directed acyclic graph technology,state channels,and shardingmechanisms,inwhich sharding emerges as a potential scaling ***,excessive cross-shard transactions and uneven shard workloads prevent the sharding mechanism from achieving the expected *** paper proposes a graphbased sharding scheme for public blockchain to efficiently balance the transaction *** crossshard transactions and evening-out workloads among shards,the scheme reduces transaction confirmation latency and enhances the transaction capacity of the ***,the scheme can achieve a high-frequency transaction as well as a better blockchain *** results show that the scheme effectively reduces the cross-shard transaction ratio to a range of 35%-56%and significantly decreases the transaction confirmation latency to 6 s in a blockchain with no more than 25 shards.

关键词： Blockchain sharding graph partitioning algorithm

来源：评论

学校读者我要写书评

暂无评论

Effective Integration of Text Diffusion and Pre-Trained Language Models with Linguistic Easy-First Schedule 30

Effective Integration of Text Diffusion and Pre-Trained Lang...

引用

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Ou, Yimin Jian, Ping School of Computer Science and Technology Beijing Institute of Technology Beijing China Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Applications Beijing Institute of Technology Beijing China

ISBN: (纸本)9782493814104

Diffusion models have become a powerful generative modeling paradigm, achieving great success in continuous data patterns. However, the discrete nature of text data results in compatibility issues between continuous diffusion models (CDMs) and pre-trained language models (PLMs). That is, the performance of diffusion models even degrades when combined with PLMs. To alleviate this issue, we propose to utilize a pre-trained decoder to convert the denoised embedding vectors into natural language instead of using the widely used rounding operation. In this way, CDMs can be more effectively combined with PLMs. Additionally, considering that existing noise schedules in text diffusion models do not take into account the linguistic differences among tokens, which violates the easy-first policy for text generation, we propose a linguistic easy-first schedule that incorporates the measure of word importance, conforming to easy-first-generation linguistic features and bringing about improved generation quality. Experiment results on the E2E dataset and five controllable tasks show that our approach can combine the merits of CDMs and PLMs, significantly outperforming other diffusion-based models. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Diffusion

来源：评论

学校读者我要写书评

暂无评论

Improving Implicit Discourse Relation Recognition with Semantics Confrontation 30

Improving Implicit Discourse Relation Recognition with Seman...

引用

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Cai, Mingyang Yang, Zhen Jian, Ping School of Computer Science and Technology Beijing Institute of Technology Beijing China Beijing Engineering Research Center of High Volume Language Information Processing and Cloud Computing Applications Beijing Institute of Technology Beijing China

ISBN: (纸本)9782493814104

Implicit Discourse Relation Recognition (IDRR), which infers discourse logical relations without explicit connectives, is one of the most challenging tasks in natural language processing (NLP). Recently, pre-trained language models (PLMs) have yielded impressive results across numerous NLP tasks, but their performance still remains unsatisfactory in IDRR. We argue that prior studies have not fully harnessed the potential of PLMs, thereby resulting in a mixture of logical semantics, which determine the logical relations between discourse arguments, and general semantics, which encapsulate the non-logical contextual aspects (detailed in Sec.1). Such a mixture would inevitably compromise the logic reasoning ability of PLMs. Therefore, we propose a novel method that trains the PLMs through two semantics enhancers to implicitly differentiate logical and general semantics, ultimately achieving logical semantics enhancement. Due to the characteristic of PLM in word representation learning, these two semantics enhancers will inherently confront with each other, facilitating an augmentation of logical semantics by disentangling them from general semantics. The experimental results on PDTB 2.0 dataset show that the confrontation approach exceeds our baseline by 3.81% F1 score, and the effectiveness of the semantics confrontation method is validated by comprehensive ablation experiments. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Semantics

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：