检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Wang, Yixuan Wang, Baoxin Liu, Yijun Wu, Dayong Che, Wanxiang Research Center for Social Computing and Information Retrieval Harbin Institute of Technology China State Key Laboratory of Cognitive Intelligence iFLYTEK Research China

Over-correction is a critical problem in Chinese grammatical error correction (CGEC) task. Recent work using model ensemble methods based on voting can effectively mitigate over-correction and improve the precision of the GEC system. However, these methods still require the output of several GEC systems and inevitably lead to reduced error recall. In this light, we propose the LM-Combiner, a rewriting model that can directly modify the over-correction of GEC system outputs without a model ensemble. Specifically, we train the model on an over-correction dataset constructed through the proposed K-fold cross inference method, which allows it to directly generate filtered sentences by combining the original and the over-corrected text. In the inference stage, we directly take the original sentences and the output results of other systems as input and then obtain the filtered sentences through LM-Combiner. Experiments on the FCGEC dataset show that our proposed method effectively alleviates the over-correction of the original system (+18.2 Precision) while ensuring the error recall remains unchanged. Besides, we find that LM-Combiner still has a good rewriting performance even with small parameters and few training data, and thus can cost-effectively mitigate the over-correction of black-box GEC systems (e.g., ChatGPT). Copyright © 2024, The Authors. All rights reserved.

关键词： Error correction

来源：评论

学校读者我要写书评

暂无评论

Insights into Psychrophilic Extracellular Electron Transfer Based on Single-Cell Metagenomics and Metagenomics Approaches

SSRN

引用

SSRN 2022年

作者： Yang, Yang Fang, Anran Feng, Kun Zhou, Huihui Xing, Defeng School of Environment State Key Laboratory of Urban Water Resource and Environment Harbin Institute of Technology Harbin150090 China Heilongjiang Province Key Laboratory of Environmental Biotechnology Harbin Institute of Technology Harbin150090 China

Revealing the metabolic functions and the extracellular electron transfer (EET) process of psychrophilic electroactive biofilms (PEB) is important for understanding the functional adaptability of electroactive bacteria to low temperatures. In this study, single cell sorting (SCS)-based metagenomics sequencing was used to characterize the composition and function of the PEB. Biofilm microbiome analysis showed that Geobacter , with a relative abundance of 62.64%, dominated the PEB enriched in the bioelectrochemical system (BES) at 4°C. Both the metagenome (MAG) and single-cell metagenome (SCM) revealed that there were no obvious metabolic differences between the original biofilm and the sorted single cells. The presence of genes associated with type IV pilus, c-type cytochromes, and riboflavin indicated that the EET potential was maintained in the PEB at low temperatures. These results suggested that SCM provides an alternative approach to reconstruct the metabolic functions of uncultured and slow-growing electroactive bacteria. © 2022, The Authors. All rights reserved.

关键词： Bacteria

来源：评论

学校读者我要写书评

暂无评论

Deep-Learning-Based Bridge Condition Assessment by Probability Density Distribution Reconstruction of Girder Vertical Deflection and Cable Tension Using Unsupervised Image Transformation Model 10th

Deep-Learning-Based Bridge Condition Assessment by Probabili...

引用

10th European Workshop on Structural Health Monitoring, EWSHM 2020

作者： Xu, Yang Tian, Yadi Zhang, Yufeng Li, Hui Key Lab of Smart Prevention and Mitigation of Civil Engineering Disasters of the Ministry of Industry and Information Technology Harbin Institute of Technology Harbin150090 China Key Lab of Structures Dynamics Behavior and Control of the Ministry of Education Harbin Institute of Technology Harbin150090 China School of Civil Engineering Harbin Institute of Technology Harbin150090 China State Key Laboratory of Safety and Health for Inservice Long Span Bridges Nanjing211112 China

ISBN: (纸本)9783030649074

This study proposes a deep-learning-based condition assessment approach by reconstructing marginal probability density distributions of girder vertical deflection (GVD) and cable tension (CT) using unsupervised image transformation (UNIT) model for a real cable-stayed bridge. 27 and 139 sensors of GVD and CT are distributed along the full bridge, and both the sampling frequencies are 2 Hz. First, vehicle-induced components of GVD and CT are extracted by data pre-processing while the temperature-induced components are removed, a time window of 3 h with a sliding length of 10 min is used to obtain original GVD and CT segments, and the corresponding margin probability density distributions (PDFs) are calculated by kernel density estimation. The training and test sets consist of 2986 and 6236 PDFs, respectively. The proposed UNIT model consists of a series of variational autoencoders (VAEs) and generative adversarial networks (GANs). Both the PDFs of GVD and CT are used as inputs. Then, the proposed UNIT model is updated by solving the mini-max problem of training GANs, in which VAEs act as generative models. Finally, the Wasserstein distance between the predicted and ground-truth PDFs of CT acts as the indicator of bridge condition change. Results show that specific modes of PDF variations induced by SHM system upgrade, cable damage, data anomaly, and traffic jam can be recognized to assess bridge condition. © 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

An Intelligent Drill Pipe Communication Approach Based on VMSK Modulation

An Intelligent Drill Pipe Communication Approach Based on VM...

引用

Computer Applications and Information technology (ISCAIT), International Symposium on

作者： Jingang Jiao Yi Wu Bingxuan Li Renjun Xie Hui Li Shiping Zhang Deng Yue State Key Laboratory of Offshore Oil Exploitation China National Offshore Oil Corporation Research Institute Beijing China China Oilfield Services Limited Beijing China School of Instrumentation Science and Engineering Harbin Institute of Technology Harbin China

ISBN: (数字)9798331542856

ISBN: (纸本)9798331542863

To address the trade-off between communication rate and transmission distance in intelligent drill pipe systems, this paper proposes a novel communication scheme based on VMSK (Very Minimum Shift keying) technology. The VMSK technique modulates carrier waves with distinct positive and negative half-cycle waveforms, enabling the differentiation of binary information (0 and 1) through the analysis of waveform variations between these half-cycles. This innovative approach allows the representation of one bit of information within a single carrier cycle, significantly enhancing data transmission rates. Moreover, we develop a specialized VMSK-based communication scheme, which conducts comprehensive simulation analysis of VMSK’s time-domain waveforms and frequency spectrum characteristics, and combines with the unique single-resonance-frequency transmission properties of intelligent drill pipes The proposed approach has been experimentally validated in an intelligent drill pipe simulation system, demonstrating the feasibility and effectiveness of the VMSK communication approach.

关键词： Time-frequency analysis Protocols Frequency modulation Resonant frequency Digital communication Time-domain analysis Information technology Standards Testing Demodulation

来源：评论

学校读者我要写书评

暂无评论

Design and Verification of a Projectile Structure for Main-belt Comets Penetration Exploration

Design and Verification of a Projectile Structure for Main-b...

引用

IEEE International Conference on robotics and Biomimetics

作者： Junyue Tang He Liang Cheng Chi Junxiao Xiao Zixiao Lu Ziyun Yao Zekun Ha Shengyuan Jiang State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin P. R. China National Center for Nanoscience and Technology CAS Center for Excellence in Nanoscience Beijing P. R. China

ISBN: (纸本)9781665481106

Comparing the isotopic composition of the D/H ratio of the water ice in Main-belt Comets (MBCs) to that of the oceans on the Earth can enhance the understanding of where the water comes from and of how the distribution of the water ices in our early solar system. Herein, a jet penetrator with a turbo-blasting explosion device and a high-precision accelerometer has been proposed to realize a 1m class subsurface detection for China next asteroidal exploration mission. Considering the ruggedness of rubble pile structure in the low-gravity environment, the projectile's posture and its overloads determined by its structure and velocity during the penetration is highly uncertain. By establishing the rubble pile structure penetrating mechanism model, the real-time overloads and posture is analyzed to acquire an optimized projectile structure. Finally, a frozen rubble pile simulant penetrating experiment under gas gun system verified the optimized structure.

关键词： Comets Analytical models Projectiles Oceans Ice Real-time systems Solar system

来源：评论

学校读者我要写书评

暂无评论

Cervical vertebra tractor: Current status and future perspectives

引用

Recent Patents on Mechanical Engineering 2020年第3期13卷 242-255页

作者： Jiang, Jingang Qian, Wei Huang, Zhiyuan Zhang, Yongde Chen, Houjun Song, Shichang Key Laboratory of Advanced Manufacturing and Intelligent Technology Ministry of Education Harbin University of Science and Technology Harbin150080 China Robotics & ITS Engineering Research Center Harbin University of Science and Technology Harbin150080 China School of Foreign Language Harbin Institute of Technology Harbin150001 China

Background: According to "The World's Top Ten Chronic Diseases" published by the World Health Organization (WHO), Cervical Spondylosis, ranked second, has afflicted more youngsters and the number is still growing because of the popularity of smart devices and long-term office work. Surgical treatment is an invasive method with certain harm;in contrast, traction therapy, as a conservative treatment, can effectively relieve neck pain and oppressed nerves, which is widely accepted by doctors and patients. How to effectively guide and implement traction therapy has become a research hotspot in the field of cervical rehabilitation. Objective: To provide an overview of the existing cervical vertebra tractors and introduce their classification, characteristics, and development. Methods: This paper reviews the principles of traction therapy and various patents related to the cervical vertebra tractor. The structural characteristics, differentiations, and applications of existing cervical vertebra tractors are also introduced. Results: The existing cervical vertebra tractors are analyzed and compared, and the typical characteristics are concluded. The main problems in its development are analyzed, the development trend is foreseen, and the current and future research of the productions and patents on the cervical vertebra tractor are discussed. Conclusion: The cervical vertebra tractor is divided into a sitting and supine device according to the application of traction postural. The device can relieve the nerve, blood vessel, and spinal cord, relieve pain, and quickly relieve the symptoms of cervical spondylosis. However, the current treatment device exerts single axial traction, and the normal physiological curvature of the human cervical vertebra is similar to the shape of "C", which is not conducive to maintaining normal physiological curvature. © 2020, Bentham Science Publishers. All rights reserved.

关键词： Patient treatment

来源：评论

学校读者我要写书评

暂无评论

WSCFER: Improving Facial Expression Representations by Weak Supervised Contrastive Learning

WSCFER: Improving Facial Expression Representations by Weak ...

引用

IEEE/RSJ International Conference on Intelligent Robots and systems (IROS)

作者： Wei Nie Bowen Chen Wenhao Wu Xiu Xu Weihong Ren Honghai Liu State Key Laboratory of Robotics and Systems Harbin Institute of Technology Shenzhen Shenzhen China Department of Child Health Care Children's Hospital of Fudan University Shanghai China

The major challenge of Facial Expression Recog-nition (FER) is to learn class discriminative representations, and the existing works mainly address it by designing various classification networks from class level. However, learning representations at class level is limited due to the inconspicuous class discrimination among different facial expressions. Thus, in this paper, we propose a Weak Supervised Contrastive learning FER (WSCFER) method to improve facial expression representations by simultaneously learning instance-level representations which are highly complementary to the general class-level representations. Specifically, our proposed WSCFER consists of three components: a major task for FER classification, an auxiliary task for Weak Supervised Contrastive (WSC) learning which pulls augmented samples of the same image together while pushing apart instance samples from different classes, and a Partial Consistency Loss (PCL) for optimizing the two embedding spaces from both the class level and the instance level. We compare WSC with some state-of-the-art contrastive methods and find that it can efficiently learn instance-level representations but avoid overemphasizing irrelevant parts, which is crucial for FER. WSCFER achieves superior performance on several in-the-wild databases, and it also shows the promising potential for learning representations under noisy annotations.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Modeling of the Constant-Current Stimuli Response of a Bio-Robot for Long-Term Motion Control

Modeling of the Constant-Current Stimuli Response of a Bio-R...

引用

IEEE International Conference on robotics and Biomimetics

作者： Jie Zhou Yuan Fang Yang Chen Yao Li Bing Li Guangdong Provincial Key Laboratory of Intelligent Morphing Mechanisms and Adaptive Robotics Key University Laboratory of Mechanism & Machine Theory and Intelligent Unmanned Systems of Guangdong and with School of Mechanical Engineering and Automation Harbin Institute of Technology Shenzhen China

ISBN: (数字)9798331509644

ISBN: (纸本)9798331509651

Bio-robots continue to receive attention because of their small size and low power consumption. Unfortunately, it currently faces two major obstacles before practical application. The first is the habituation issue, insects are difficult to be effectively controlled for a long time by electrical stimuli signals. The other one is the lack of model for motion control. In this research, we established a stimuli-response model for cockroach bio-robot based on long time current. A more convenient optic lobe implantation method has been proposed, which has lower attenuation and better stimuli effect. Constant electrical stimuli also significantly increased the number of effective controls for crawling. A cockroach bio-robot stimuli distance response model was established and validated with experiment data. Five parameters were including in this model, namely sex, length, weight, current amplitude and equivalent stiumated-number. This model can predict the movement distance of biological robots well.

关键词： Power demand Biological system modeling Insects Optical attenuators Predictive models Brain modeling Data models Motion control Robots Faces

来源：评论

学校读者我要写书评

暂无评论

Research on UAV Detection technology of Gm-APD Lidar Based on YOLO Model

Research on UAV Detection Technology of Gm-APD Lidar Based o...

引用

2021 IEEE International Conference on Unmanned systems, ICUS 2021

作者： DIng, Yuanxue Qu, Yanchen Zhang, Qing Tong, Jiahui Yang, Xianhui Sun, Jianfeng Harbin Institute of Technology National Key Laboratory of Science and Technology on Tunable Laser Institution of Opto-Electronics Harbin China Beijing Electromechanical Engineering Institute Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Beijing China

ISBN: (纸本)9780738146577

In recent years, unmanned aerial vehicle (UAV) technology has developed rapidly, which plays an important role in both military and civil fields. While it brings convenience to all walks of life, there are also a lot of security problems. Therefore, it is necessary to study anti-UAV technology, but the detection of UAV is an important basis of anti-UAV technology. In this paper, by combining the three-dimensional range profile collected by Gm-APD (Geiger mode Avalanche Photo Diode) lidar with the deep learning method, a UAV detection method based on the improved YOLOv3 (You Only Look Once v3) network is proposed. Firstly, the data of UAV in low altitude real scene is collected by Gm-APD lidar, and range profile information is generated. Then, the YOLOv3 network is improved, and the SPP module can be added to combine the local features of the target with the global features to improve the detection accuracy. It is suitable for the detection of UAVs. Finally, the trajectory of the UAV is drawn based on the distance information. The experimental results show that the improved method can effectively detect UAV targets, and is better than the previous method, YOLOv3-Tiny method and YOLOv4 method. The farthest detection distance is 238m, the detection speed is fast, and the purpose of real-time detection and positioning can be achieved. Therefore, this method can be applied to anti-UAV technology and has important research significance. © 2021 IEEE.

关键词： Optical radar

来源：评论

学校读者我要写书评

暂无评论

一体化可拉伸的多功能电化学传感纤维构建微量汗液健康监测织物

引用

Science China Materials 2024年第1期67卷 251-260页

作者：张安宁周靓靓梁琦敏王欣胡晓康贾康康储宏伟罗勇锋丘龙斌彭慧胜何思斯 Shenzhen Key Laboratory of Flexible Printed Electronics Technology School of ScienceHarbin Institute of Technology(Shenzhen)Shenzhen 518055China Department of Mechanical and Energy Engineering Southern University of Science and TechnologyShenzhen 518055China Hunan Province Key Laboratory of Materials Surface&Interface Science and Technology College of ScienceMaterial Science and Engineering SchoolCentral South University of Forestry and TechnologyChangsha 410004China State Key Laboratory of Molecular Engineering of Polymers Department of Macromolecular Science and Laboratory of Advanced MaterialsFudan UniversityShanghai 200438China

透气舒适的传感纺织品可以检测人体汗液中的多种生物标记物,是在日常生活中实现全面健康监测的有效途径.然而,目前的可穿戴柔性电化学纺织品缺乏伸展性,这可能会导致信号不稳定或在移动过程中损坏设备.此外,这些纺织品对多种指示器的集... 详细信息

透气舒适的传感纺织品可以检测人体汗液中的多种生物标记物,是在日常生活中实现全面健康监测的有效途径.然而,目前的可穿戴柔性电化学纺织品缺乏伸展性,这可能会导致信号不稳定或在移动过程中损坏设备.此外,这些纺织品对多种指示器的集成度有限,需要较大的表面积和大量的汗液来激活传感器.在此,我们报告了一种一体化多功能电化学生物传感器纤维,该纤维具有可拉伸性,通过微量汗液即可检测汗液中的多种生物标志物.这种生物传感器是通过将多功能碳纳米管条带以螺旋状排列的方式负载在预先拉伸的聚合物纤维芯上从而充当微电极,并且两者之间具有稳定界面.此外,通过引入超亲水鞘层,提高了生物传感器的汗液捕获效率.该生物传感器能够同时监测pH、K^(+)、Na^(+)、葡萄糖、乳酸和尿酸六种生物标记物,300%的应变下表现出稳定的传感性能.仅需1μL的汗液即可启动对六种生物标记物进行高效检测.由此编织的织物传感系统可以连续、实时地监测多种生物标记信息,从而完成人体健康状况的评估.

关键词：生物传感器生物标记物人体健康状况健康监测聚合物纤维损坏设备伸展性微电极

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：