检索结果-内蒙古大学图书馆

Knowledge-Empowered,Collaborative,and Co-Evolving AI Models:The Post-LLM Roadmap

工程（英文） 2025年第1期44卷 87-100页

作者： Fei Wu Tao Shen Thomas Bäck Jingyuan Chen Gang Huang Yaochu Jin Kun Kuang Mengze Li Cewu Lu Jiaxu Miao Yongwei Wang Ying Wei Fan Wu Junchi Yan Hongxia Yang Yi Yang Shengyu Zhang Zhou Zhao Yueting Zhuang Yunhe Pan College of Computer Science and Technology Zhejiang UniversityHangzhou 310027China Leiden Institute of Advanced Computer Science Leiden UniversityLeiden 2333 CCNetherlands College of Electrical Engineering Zhejiang UniversityHangzhou 310027China School of Engineering Westlake UniversityHangzhou 310024China Department of Computer Science and Engineering Hong Kong University of Science and TechnologyHong Kong 999077China Department of Computer Science and Engineering Shanghai Jiao Tong UniversityShanghai 200240China School of Cyber Science and Technology Sun Yat-Sen UniversityShenzhen 518107China Department of Computing The Hong Kong Polytechnic UniversityHong Kong 999077China

Large language models(LLMs)have significantly advanced artificial intelligence(AI)by excelling in tasks such as understanding,generation,and reasoning across multiple *** these achieve-ments,LLMs have inherent limitations including outdated information,hallucinations,inefficiency,lack of interpretability,and challenges in domain-specific *** address these issues,this survey explores three promising directions in the post-LLM era:knowledge empowerment,model collaboration,and model ***,we examine methods of integrating external knowledge into LLMs to enhance factual accuracy,reasoning capabilities,and interpretability,including incorporating knowledge into training objectives,instruction tuning,retrieval-augmented inference,and knowledge ***,we discuss model collaboration strategies that leverage the complementary strengths of LLMs and smaller models to improve efficiency and domain-specific performance through techniques such as model merging,functional model collaboration,and knowledge ***,we delve into model co-evolution,in which multiple models collaboratively evolve by sharing knowledge,parameters,and learning strategies to adapt to dynamic environments and tasks,thereby enhancing their adaptability and continual *** illustrate how the integration of these techniques advances AI capabilities in science,engineering,and society—particularly in hypothesis development,problem formulation,problem-solving,and interpretability across various *** conclude by outlining future pathways for further advancement and applications.

关键词： Artificial intelligence Large language models Knowledge empowerment Model collaboration Model co-evolution

来源：评论

学校读者我要写书评

暂无评论

PLDMLT:Multi-Task Learning of Diabetic Retinopathy Using the Pixel-Level Labeled Fundus Images

引用

computers, Materials & Continua 2023年第8期76卷 1745-1761页

作者： Hengyang Liu Chuncheng Huang Department of Computer Science and Engineering Chongqing University of TechnologyChongqing400054China

In the field of medical images,pixel-level labels are time-consuming and expensive to acquire,while image-level labels are relatively easier to ***,it makes sense to learn more information(knowledge)from a small number of hard-to-get pixel-level annotated images to apply to different tasks to maximize their usefulness and save time and training *** this paper,using Pixel-Level Labeled Images forMulti-Task Learning(PLDMLT),we focus on grading the severity of fundus images for Diabetic Retinopathy(DR).This is because,for the segmentation task,there is a finely labeled mask,while the severity grading task is without classification *** this end,we propose a two-stage multi-label learning weakly supervised algorithm,which generates initial classification pseudo labels in the first stage and visualizes heat maps at all levels of severity using Grad-Cam to further provide medical interpretability for the classification task.A multitask model framework with U-net as the baseline is proposed in the second stage.A label update network is designed to alleviate the gradient balance between the classification and segmentation *** experimental results show that our PLDMLTmethod significantly outperforms other stateof-the-art methods in DR segmentation on two public datasets,achieving up to 98.897%segmentation *** addition,our method achieves comparable competitiveness with single-task fully supervised learning in the DR severity grading task.

关键词： DR lesion segmentation pseudo labels grading task class activation heat map update label network

来源：评论

学校读者我要写书评

暂无评论

Learner’s intention analysis to mitigate the cold start problem in personalized learning recommendation systems

引用

Multimedia Tools and Applications 2024年 1-55页

作者： Pal, Saurabh Pramanik, Pijush Kanti Dutta Choudhury, Prasenjit Department of Computer Science and Engineering School of Engineering Brainware University Kolkata India School of Computer Applications and Technology Galgotias University Uttar Pradesh Greater Noida India Department of Computer Science & amp Engineering National Institute of Technology Durgapur West Bengal India

Understanding the learner’s requirements and status is important for recommending relevant and appropriate learning materials to the learner in personalized learning. For this purpose, the learning recommendation system must include the learner’s contextual and behavioral information. However, in the case of cold start problems, the system has no prior information about a new user. Moreover, the system cannot acquire the required information instantly because the automated and instantaneous acquisition of this information is not trivial. This hinders the personalized recommendation for a new learner. Engaging the learner in a questionnaire session may allow the learning system to acquire most of the required information. However, this makes the learner uninterested and disengaged with the system. To address this, we propose a recommendation system for personalized learning that mitigates the cold start problem by acquiring important information by analyzing the learner’s questions submitted to the learning system. By analyzing the meta-information of a question, we identified the learner’s exact requirements, domain and topic knowledge, and expected knowledge gain from the learning material. Ontology, BiLSTM, and context mapping are used for learner modelling and topic classification, and identifying suitable learning materials, respectively. Furthermore, MCDM techniques are used for ranking the learning materials. Our experimental assessment and validation suggest that the information acquired based on our proposed approach is sufficient for understanding a new learner’s needs and suitability and, accordingly, for recommending suitable personalized learning materials. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Comprehensive analysis of various imputation and forecasting models for predicting PM2.5 pollutant in Delhi

引用

Neural Computing and Applications 2025年第17期37卷 11441-11458页

作者： Karnati, Hemanth Soma, Anuraag Alam, Adnan Kalaavathi, B. School of Computer Science and Engineering Vellore Institute of Technology Tamil Nadu Vellore632014 India Department of IoT School of Computer Science and Engineering Vellore Institute of Technology Tamil Nadu Vellore632014 India

Understanding and predicting air quality is pivotal for public health and environmental management, especially in urban areas like Delhi. This study utilizes a comprehensive dataset from the Central Pollution Control Board, detailing air pollutant concentrations at the ITO (Income Tax Office), Delhi station, from 2017 to 2023. Our focus on pollutants such as PM2.5, PM10, NO2, NH3, SO2, ozone, and CO highlights the challenges posed by missing data in environmental studies. Through the application of several imputation models, including KNN, linear regression, forward fill + backward fill, Fourier + KNN, linear interpolation, and statistical methods, we identified combination of forward fill and backward fill as the most effective method for addressing data gaps, specifically for ozone measurements. Building on this imputed dataset, we employ various forecasting models to predict PM2.5 levels, a critical pollutant. Our exploration includes time series and deep learning approaches, such as LSTM, Bi-LSTM, CNN-LSTM, GRU, Deep AR, WaveNet, TCN, ARIMA, SARIMA, and Prophet. The performance of these models is evaluated using daily data, with Bi-LSTM and LSTM with attention emerging as the top performers due to their accuracy in predicting PM2.5 concentrations. The implications of our findings extend beyond academic interest, offering practical insights for environmental policy and health advisories in Delhi. By demonstrating the efficacy of specific imputation and forecasting techniques, our research contributes to the broader effort of improving air quality monitoring and prediction. The success of Bi-LSTM and LSTM with attention models, in particular, suggests a promising avenue for future studies aiming to enhance the precision of environmental forecasts. © The Author(s) 2025.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

Vision-based Robotic Arm in Defect Detection and Object Classification Applications

引用

Sensors and Materials 2024年第2期36卷 655-670页

作者： Lin, Cheng-Jian Jhang, Jyun-Yu Gao, Yi-Jyun Huang, Hsiu-Mei Department of Computer Science & Information Engineering National Chin-Yi University of Technology Taichung411 Taiwan Department of Computer Science and Information Engineering National Taichung University of Science and Technology Taichung404 Taiwan Department of Information Management National Taichung University of Science and Technology Taichung404 Taiwan

Robotic arms have been widely used in industrial fields. However, researchers have seldom considered the factors affecting the actual factory environment. For example, when objects are conveyed in a factory, conveyor belts are often used to dynamically plan the overall production line. In addition, each object requires multiple checkpoints for repeated audits and inspections to ensure its quality. In this study, a vision-based robotic arm system equipped with multiple functionalities was developed. The development process consisted of three steps: detecting multiple dynamic objects, determining the size of each object, and identifying object defects. In the first step, You Only Look Once was used to detect multiple dynamic objects on a conveyor belt in real time. In the second step, the original image of the object was converted into a grayscale image, and the edge contour of the object was drawn using a Canny edge detection algorithm. Objects in the image were then rotated for vertical and horizontal projections, and then an artificial neural network (ANN) was used to calculate the size of each object. In the third step, a convolutional fuzzy neural network (CFNN) was used to identify object defects. This network was divided into an input layer, a convolution pooling layer, a feature fusion layer, a fuzzy layer, a regular layer, and a defuzzification layer. According to the experimental results, the standard error of the mean between the object size obtained by the ANN and the actual size was 0.009. In addition, the accuracy, recall, precision, and F1-score obtained by the CFNN in object defect detection were 0.9580, 0.9535, 0.9535, and 0.9535, respectively. Compared with other deep neural network models, such as AlexNet and LeNet, the proposed CFNN has fewer parameters and higher performance. © MYU K.K.

关键词： Fuzzy neural networks

来源：评论

学校读者我要写书评

暂无评论

Deep-BERT:Transfer Learning for Classifying Multilingual Offensive Texts on Social Media

引用

computer Systems science & engineering 2023年第2期44卷 1775-1791页

作者： Md.Anwar Hussen Wadud M.F.Mridha Jungpil Shin Kamruddin Nur Aloke Kumar Saha Department of Computer Science and Engineering Bangladesh University of Business and TechnologyDhakaBangladesh School of Computer Science and Engineering University of AizuAizuwakamatsuJapan Department of Computer Science American International University-BangladeshDhakaBangladesh Department of Computer Science and Engineering University of Asia PacificDhakaBangladesh

Offensive messages on social media,have recently been frequently used to harass and criticize *** recent studies,many promising algorithms have been developed to identify offensive *** algorithms analyze text in a unidirectional manner,where a bidirectional method can maximize performance results and capture semantic and contextual information in *** addition,there are many separate models for identifying offensive texts based on monolin-gual and multilingual,but there are a few models that can detect both monolingual and multilingual-based offensive *** this study,a detection system has been developed for both monolingual and multilingual offensive texts by combining deep convolutional neural network and bidirectional encoder representations from transformers(Deep-BERT)to identify offensive posts on social media that are used to harass *** paper explores a variety of ways to deal with multilin-gualism,including collaborative multilingual and translation-based ***,the Deep-BERT is tested on the Bengali and English datasets,including the different bidirectional encoder representations from transformers(BERT)pre-trained word-embedding techniques,and found that the proposed Deep-BERT’s efﬁcacy outperformed all existing offensive text classiﬁcation algorithms reaching an accuracy of 91.83%.The proposed model is a state-of-the-art model that can classify both monolingual-based and multilingual-based offensive texts.

关键词： Offensive text classiﬁcation deep convolutional neural network(DCNN) bidirectional encoder representations from transformers(BERT) natural language processing(NLP)

来源：评论

学校读者我要写书评

暂无评论

Study Influencing Factors of Maternal Health and the Role of Internet of Things (IoT) to Improve Maternal Care

引用

SN computer science 2024年第6期5卷 778页

作者： Rani, Deepa Kumar, Rajeev Chauhan, Naveen Department of Computer Science and Engineering National Institute of Technology

Maternal health during pregnancy is influenced by various factors that significantly impact pregnancy outcomes. This paper aims to highlight these critical factors, promote awareness, and advocate proactive self-care among women. By analyzing various works, a taxonomy of influencing factors is proposed to provide an understanding of their effects on maternal and fetal health. The taxonomy reveals multiple factors, including individual characteristics, social determinants, and environmental factors, that significantly affect maternal and fetal health. Therefore, by thoroughly reviewing reputed articles in the field of IoT technology in maternal care, a list of sensors and devices capable of capturing health-related parameters and various applications of IoT in maternal care is presented. IoT technology and promoting self-care can enhance maternal care and improve outcomes for both mother and fetus. Therefore, this research emphasizes the factors influencing pregnancy and the benefits of the adoption of IoT technology to enable a patient-centric approach for prompt detection of potential risks in maternal care. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： IoT Maternal care Remote monitoring Risk monitoring Self-care

来源：评论

学校读者我要写书评

暂无评论

Comparing HAProxy Scheduling Algorithms during the DDoS Attacks

IEEE Networking Letters

引用

IEEE Networking Letters 2024年第2期6卷 139-142页

作者： Kumar, Anmol Somani, Gaurav Agarwal, Mayank Indian Institute of Technology Patna Department of Computer Science and Engineering Patna800013 India Central University of Rajasthan Department of Computer Science and Engineering Ajmer305817 India

Web-services have become most common IT enablers today. Cyber attacks such as the distributed denial of service (DDoS) attacks pose availability concerns which may result into service outages and consequently financial and reputation losses. Load balancing software such as HAProxy (high availability proxy) is an important building block of the Web service delivery today. Load balancing software provides incoming request distribution among Web-servers and admission control which may even help with combating DDoS attacks. In this letter, we study and explore the performance of different load distribution or scheduling strategies in the presence of DDoS attacks. With the help of attack experiments, we find that First, HDR, Source, and URI scheduling algorithms performs best for both static and Poisson arrivals. © 2019 IEEE.

关键词： Scheduling algorithms

来源：评论

学校读者我要写书评

暂无评论

Resilient TCP Variant Enabling Smooth Network Updates for Software-Defined Data Center Networks

引用

Tsinghua science and technology 2024年第5期29卷 1615-1632页

作者： Abdul Basit Dogar Sami Ullah Yiran Zhang Hisham Alasmary Muhammad Waqas Sheng Chen Department of Computer Science and Technology Tsinghua UniversityBeijing 100084China and also with Department of Informatics and SystemsUniversity of Management and TechnologyLahore 54660Pakistan Department of Computer Science Shaheed Benazir Bhutto UniversitySheringalUpper Dir 18050Pakistan School of Computer Science Beijing University of Posts and TelecommunicationsBeijing 100876China Department of Computer Science College of Computer ScienceKing Khalid UniversityAbha 61421Saudi Arabia Department of Computer Engineering Faculty of Information TechnologyUniversity of BahrainSakheer 32038Bahrainand also with School of EngineeringEdith Cowan UniversityPerth WA 6027Australia School of Electronics and Computer Science University of SouthamptonSouthamptonSO171BJUK

Network updates have become increasingly prevalent since the broad adoption of software-defined networks(SDNs)in data *** TCP designs,including cutting-edge TCP variants DCTCP,CUBIC,and BBR,however,are not resilient to network updates that provoke flow *** this paper,we first demonstrate that popular TCP implementations perform inadequately in the presence of frequent and inconsistent network updates,because inconsistent and frequent network updates result in out-of-order packets and packet drops induced via transitory congestion and lead to serious performance *** look into the causes and propose a network update-friendly TCP(NUFTCP),which is an extension of the DCTCP variant,as a *** are used to assess the proposed *** findings reveal that NUFTCP can more effectively manage the problems of out-of-order packets and packet drops triggered in network updates,and it outperforms DCTCP considerably.

关键词： software defined data center networks network updates DCTCP out-of-order packets packet drop SDN

来源：评论

学校读者我要写书评

暂无评论

Extreme learning machine to foretell the deflection in plain cement, steel-reinforced, and bamboo-reinforced concrete beams

引用

Emerging Materials Research 2025年第1期14卷 12-27页

作者： Dubey, Saurabh Gupta, Deepak Mallik, Mainak Department of Civil Engineering National Institute of Technology Arunachal Pradesh India Department of Computer Science and Engineering Motilal Nehru National Institute of Technology Allahabad India

In mountainous regions, bamboo has gained popularity as a local reinforcement material for concrete due to its availability and cost-effectiveness. However, the lack of standardized design guidelines complicates the assessment of deflection behavior under varying loads. Challenges such as bond failure, shrinkage, corrosion, and material strength uncertainties further limit its structural application. To address these issues, this study employs an extreme learning machine to predict deflection in plain cement concrete beams, steel-reinforced cement concrete beams, and bamboo-reinforced concrete beams. The performance of extreme learning machine is compared with support vector regression and artificial neural networks. A total of 122 specimens were analyzed using input features such as cement content, water-cement ratio, fine aggregate, coarse aggregate, tensile strength, concrete compressive strength, percentage of reinforcement, applied load, cross-sectional area, and clear cover. The extreme learning machine method demonstrated satisfactory performance with a coefficient of deformation of 0.9975, Mean Square Error of 0.0003, Mean Absolute Error of 0.0052, Mean Absolute Percentage Error of 0.0998, and Root Mean Square Error of 0.01694. These results suggest that extreme learning machine is an efficient machine learning algorithm for predicting the deflection behavior of different reinforced concrete beams, providing a reliable alternative to traditional methods. © 2025 Emerald Publishing Limited: All rights reserved.

关键词： Mean square error

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：