检索结果-内蒙古大学图书馆

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

作者： Ranaldi, Leonardo Pucci, Giulia Freitas, Andrè Idiap Research Institute Martigny Switzerland Department of Computer Science University of Manchester United Kingdom Department of Computing Science University of Aberdeen United Kingdom

ISBN: (纸本)9782493814104

Curriculum Learning (CL) has been emerged as an effective technique for improving the performances and reducing the cost of pre-training Large Language Models (LLMs). The efficacy of CL demonstrated in different scenarios is in the training LLMs by organizing examples from the simplest to the most complex. Although improvements have been shown extensively, this approach was used for pre-training, leaving novel fine-tuning approaches such as instruction-tuning unexplored. In this paper, we propose a novel complexity measure to empower the instruction-tuning method using the CL paradigm. To complement previous works, we propose cognitively motivated measures to determine the complexity of training demonstrations used in the instruction-tuning paradigm. Hence, we experiment with the proposed heuristics first in English and then in other languages. The downstream results show that delivering training examples by complexity ranking is also effective for instruction tuning, as it improves downstream performance while reducing costs. Furthermore, the technique can be easily transferred to languages other than English, e.g., Italian and French, without any adaptation, maintaining functionality and effectiveness. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Curricula

来源：评论

学校读者我要写书评

暂无评论

Leveling Up Together: Fostering positive Growth and Safe Online Spaces for Teen Roblox Developers 25

Leveling Up Together: Fostering Positive Growth and Safe Onl...

引用

2025 CHI Conference on Human Factors in computing Systems, CHI 2025

作者： Choi, Yubin Choi, Jeanne Seering, Joseph Computer Science Korea Advanced Institute of Science & Technology Daejeon Korea Republic of School of Computing Korea Advanced Institute of Science & Technology Daejeon Korea Republic of

ISBN: (纸本)9798400713941

Creating games together is both a playful and effective way to develop skills in computational thinking, collaboration, and more. However, game development can be challenging for younger developers who lack formal training. While teenage developers frequently turn to online communities for peer support, their experiences may vary. To better understand the benefits and challenges teens face within online developer communities, we conducted interviews with 18 teenagers who created games or elements in Roblox and received peer support from one or more online Roblox developer communities. Our findings show that developer communities provide teens with valuable resources for technical, social, and career growth. However, teenagers also struggle with inter-user conflicts and a lack of community structure, leading to difficulties in handling complex issues that may arise, such as financial scams. Based on these insights, we propose takeaways for creating positive and safe online spaces for teenage game creators. © 2025 Copyright held by the owner/author(s).

关键词： Tweets

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-based Intelligent Algorithms for Effective Transmission Authentication and Anomaly Identification in Vehicular Networks

Deep Learning-based Intelligent Algorithms for Effective Tra...

引用

2023 International Conference on Innovative computing, Intelligent Communication and Smart Electrical Systems, ICSES 2023

作者： Marappan, Sibi Marappan, Harshini Sathyabama Institute of Science and Technology Department of Computer Science and Engineering Chennai600119 India SRM Institute of Science and Technology Department of Computing Technologies Kattankulathur603203 India

ISBN: (纸本)9798350319200

Vehicular networks enable effective traffic control and security on the roads. Sustaining security and network performance depends on authentic transmission and anomaly detection. The inherent delays in conventional approaches to transmission authentication and anomaly detection limit their utility in actual-time vehicular network conditions. In this paper, we present the Deep Attentive Optimized Bidirectional Long/Short Term Memory (DABLSTM) method. The method combines the advantages of deep learning with those of attention mechanisms and bidirectional LSTM. The dataset was gathered from the Hacking and Countermeasure Research Lab (HCRL). Min-max normalization is used for data pre-processing, and then Principal Component Analysis (PCA) is used for efficient feature extraction, which optimizes the source data for our method. The DA-BLSTM method examines the approach's efficiency by evaluating its performance achieves the accuracy (91.2%), recall rate (92.8%), precision (92.3%), and F1-score (93.4%). The DA-BLSTM prevails over traditional techniques by displaying lower latency and considerably higher accuracy in detecting proper transmissions and anomalies within vehicular network data. We present a promising approach for real-time, safe, reliable, and effective authentication and anomaly identification in vehicle networks, establishing an innovative standard for vehicular communication safety. © 2023 IEEE.

关键词： Anomaly identification deep attentive optimized Bidirectional long/short term memory (DA-BLSTM) Principal component analysis (PCA) Transmission authentication Vehicular network

来源：评论

学校读者我要写书评

暂无评论

Novel Lightweight Encryption Scheme Based on Transformation Matrix and ChaCha20 Algorithm for Secure Snap Transmission

Novel Lightweight Encryption Scheme Based on Transformation ...

引用

2024 IEEE International Conference on Intelligent Systems and Advanced Applications, ICISAA 2024

作者： Sellam, V. Kumar, M. Ranjith Sudeesh Kumar, V. Mahendira Kumar, S. Tharun Kaarthik, G.K. Vellore Institute of Technology School of Computer Science and Engineering Chennai India Amrita Vishwa Vidyapeetham Amrita School of Computing Dept. of Computer Science and Engg. Chennai India

ISBN: (纸本)9798331539948

A revolution for ultra-security in image transmission, due to the innovation and upgrade of cryptographic algorithms occurred during recent years. Proposed image encryption algorithm with ChaCha20 Our matrix-based shuffling not only improves image security but also adds to the complexity of data stealth. The scheme is made up of four main operations: Rearranging pixels in the image with a transform matrix, encryption using ChaCha20, decryption to get back the original image and finally descrambling it by applying inverse transformation changes. In this way, integrity is maintained and image privacy becomes secure. The provided transformation matrix makes scrambling with reversible operations and speed possible against all of the known standard attacks to which ChaCha20 encryption is subjected. The suggested method, led to a Peak Signal Noise Ratio of 361.20 and zero Mean Squared Error indicating a top-quality reconstructed image concerning the original image. The study is raising the security alerts provided by ChaCha20, a relatively new but efficient and high-fidelity encryption solution that makes it an attractive option in today's increasingly digitalized environment where data security has become everyone's top priority. © 2024 IEEE.

关键词： Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

An Extended Agglomerative Hierarchical Clustering Techniques 2

An Extended Agglomerative Hierarchical Clustering Techniques

引用

2nd IEEE International Conference on Advances in computing, Communication and Applied Informatics, ACCAI 2023

作者： Maravarman, M. Babu, S. Pitchai, R. Srm Institute of Science and Technology School of Computing Department of Computing Technologies Tamil Nadu Kattangulathur India B v Raju Institute of Technology Department of Computer Science and Engineering Telangana Narsapur India

ISBN: (纸本)9798350315905

Clustering is a significant method of data analytics in real world environments since human labelling of the data is often costly. Clustering was developed as an alternative to manual tagging. In the field of data analytics, hierarchical clustering is of critical significance, particularly in light of the exponential rise of data derived from the normal world. You may derive a variety of hierarchical agglomerative clustering algorithms from this architecture by providing an inter-cluster semantic similarity, an expression patterns of the -similarity graph, and a cover procedure. These three pieces of information are required. According to the findings of our experiments, our approaches are not only more efficient than conventional hierarchical algorithms, but they also produce smaller agglomerative hierarchical clustering while maintaining the same level of clustering effectiveness. It is generally agreed that topology management is an effective strategy for addressing these challenges. This method groups nodes together for the purpose of managing them and/or carrying out a variety of duties in a dispersed way, such as resource management. There are many quality-driven goals that may be accomplished by clustering, despite the fact that approaches for clustering are mostly renowned for their ability to reduce energy usage. The purpose of this study is to provide a comprehensive explanation on various enhanced agglomerative hierarchical clustering techniques. In addition to this, the authors have provided certain criteria, on the basis of which one may also assess which of these previously described algorithms is the most effective. © 2023 IEEE.

关键词： Data Analytics

来源：评论

学校读者我要写书评

暂无评论

WNet: Multi-Task Artistic Portrait Generation Method 14

WNet: Multi-Task Artistic Portrait Generation Method

引用

14th International Conference on Information Technology in Medicine and Education, ITME 2024

作者： Li, Juan Wu, Binlong Lin, Yuhui Zhang, Huang Huang, Xiangping Fuzhou Institute of Technology College of Computing and Information Science Fuzhou350506 China Fuzhou University College of Computer and Data Science Fuzhou350108 China

ISBN: (纸本)9798350356670

Artistic portrait usually uses a set of sparse continuous lines to capture a person's unique appearance characteristics based on real people or photos. This paper addresses the issues of strong background noise and the loss of key details of human faces in artistic portraiture, leading to the inability to produce realistic and high quality portraits. A multi-task portrait painting model with hierarchical loss called WNet is proposed based on U2Net used in salient object detection. This model can generate highly abstract portrait painting that do not depend on the background. Compared to existing popular portrait painting generation methods, the images generated by our method have a better visual effect and retain the key details of the facial features better. The experimental results indicate that the proposed WNet is optimal in terms of PSNR, SSIM and MAE performance. Finally, The software communicating robotic arm can quickly spot-paint the fused art portraits. © 2024 IEEE.

关键词： Robotic arms

来源：评论

学校读者我要写书评

暂无评论

Enhancing Malicious QR Code Link Detection with Swarm-Intelligent Squirrel-Search Optimized ANN

Enhancing Malicious QR Code Link Detection with Swarm-Intell...

引用

2023 International Conference on Innovative computing, Intelligent Communication and Smart Electrical Systems, ICSES 2023

ISBN: (纸本)9798350319200

Recently, hackers have made plans to replicate harmful links by tricking people in a variety of ways. By acquiring access to sensitive information they use online, they attempt to remotely manipulate the victims' machines or obtain their data. QR codes are two-dimensional barcodes that can encode different types of data and are viewable by digital devices like smart phones. However, there is no recognized procedure for creating QR codes;therefore they could be vulnerable to a number of dubious assaults. The goal of this study is to use the Swarm-Intelligent Squirrel-Search Optimized Artificial Neural Network (SISSO-ANN) to identify harmful URLs concealed behind 2D QR codes. On the basis of barcode forgery, a cyber attack that may be carried out online has been proposed. Through a number of sites, a dataset of harmful and helpful Uniform Resource Locators (URLs) was produced, and their lexical properties were gathered. The experimental finding demonstrates that SISSO-ANN works better than other current techniques and achieves detection precision value of 93.80%,and accuracy of 93.50% which is much greater than the present secure QR code scanners. Additionally, the application with the least privilege authorization, SISSO-ANN, is the most privacy-friendly. © 2023 IEEE.

关键词： Malicious Links Quick Response (QR) Swarm-Intelligent Squirrel-Search Optimized Artificial Neural Network (SISSO-ANN) Uniform Resource Locators (URLs)

来源：评论

学校读者我要写书评

暂无评论

Secured Normalized Tagged Response for Patient Critical Care Monitoring System: A Unified approach

Secured Normalized Tagged Response for Patient Critical Care...

引用

2023 International Conference on Advances in Computation, Communication and Information Technology, ICAICCIT 2023

作者： Sivakumar, Soubraylu Vinay, B. Sai Arulmurugan, A. Prakash, U.M. Kumar, T. Rajesh Sridevi, S. Srm Institute of Science and Technology Computer Technologies School of Computing Chennai Kattankulathur India Saveetha Institute of Medical and Technical Sciences Computer Science and Engineering Saveetha School of Engineering Chennai India Vels Institute of Science Technology and Advanced Studies Computer Science and Engineering Chennai India

ISBN: (纸本)9798350344387

In the rapidly advancing realm of healthcare, ensuring the well-being of critical care patients stands as a paramount mission. Various cutting-edge sensors are available in the market, proficient in monitoring vital health. A substantial number of healthcare facilities still rely on manual processes, a practice that is not only repetitive and labor-intensive but also fraught with the potential for human errors. This pressing need for immediate, accurate responses in the context of critical care has driven the development of a comprehensive solution. Our proposed system brings together security, data normalization, and efficient data exchange, underpinned by a unified approach. Security, a cornerstone of healthcare, finds its realization in the implementation of JSON Web Tokens (JWT). This technology ensures the privacy and integrity of communication between healthcare entities. Crucial patient data, including personal particulars, lifeline parameters, and historical records, is securely housed in a scalable and non-redundant Postgres SQL database. In emergencies, time is of the essence. To expedite responses, an innovative tagged QR code technique is introduced. The unified approach doesn't simply prioritize security;it underscores the significance of data standardization and efficient data exchange. A robust security framework has been meticulously designed to protect sensitive patient information, minimize the risk of data breaches, and prevent unauthorized access. Furthermore, data normalization ensures that patient data is consistently structured and formatted, facilitating compatibility with diverse healthcare platforms and devices. The normalization additionally enhances interoperability, enabling seamless information exchange between different systems and healthcare providers. In summary, our unified approach, marked by state-of-the-art security, data normalization, and efficient data tagging, sets a new standard for patient care in critical settings, ensurin

关键词： Critical Care Emergency Health Care JSON Web Token Patient monitoring Postgres SQL QR Code

来源：评论

学校读者我要写书评

暂无评论

Innovative Video Classification Method based on Deep Learning Approach

引用

Recent Patents on Engineering 2025年第2期19卷 95-107页

作者： Hemamalini, V. Jayasutha, D. Vinothini, V.R. Devi, R. Manjula Kumar, A. Anitha, E. Department of Networking and Communications School of Computing SRM Institute of Science and Technology Chennai India Department of Computer Science and Engineering Arjun College of Technology Coimbatore Thamaraikulam India Department of Mathematics Bannari Amman Institute of Technology Sathyamangalam Erode638401 India Department of Computer Science and Design Kongu Engineering College Perundurai Erode638060 India Department of Civil Engineering University Institute of Engineering & Technology Maharshi Dayanand University Rohtak124001 India Department of Artificial Intelligence and Data Science Sri Eshwar College of Engineering Coimbatore641202 India

Background: The automated classification of videos through artificial neural networks is addressed in this work. To explore the concepts and measure the results, the data set UCF101 is used, consisting of video clips taken from YouTube to recognize actions. The study is carried out with the authors' resources to determine the feasibility of independent research in the area. Methods: This work was developed in the Python programming language using the Keras library with Tensorflow as the back-end. The objective is to develop a network that presents performance compatible with the state of the art in terms of classifying videos according to the actions taken. Results: Given the hardware limitations, there is considerable distance between the implementation possibilities in this work and what is known as the state-of-the-art. Conclusion: Throughout the work, some aspects in which this limitation influenced the development are presented, but it is shown that this realization is feasible and that obtaining expressive results is possible 98.6% accuracy is obtained in the UCF101 data set, compared to the 98 percentage points of the best result ever reported, using, however, considerably fewer resources. In addition, the importance of transfer learning in achieving expressive results as well as the different performances of each architecture are reviewed. Thus, this work may open doors to carry patent-based outcomes. © 2025 Bentham science Publishers.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Enhanced Real-Time Object Detection on Customized Dataset with YOLOv9 Architecture

Enhanced Real-Time Object Detection on Customized Dataset wi...

引用

2024 IEEE International Conference on Intelligent Signal Processing and Effective Communication Technologies, INSPECT 2024

作者： Adithiyan, P.V. Kumar, M Ranjith Rahul, K. Nikhil, V. Annamalai, R. Amrita School of Computing Amrita Vishwa Vidyapeetham Dept. of Computer Science and Engg. Chennai India School of Computer Science and Engineering Vellore Institute of Technology Chennai India

ISBN: (纸本)9798350379525

YOLO has developed into a primary real-time object identification platform for applications such as video surveillance systems, autonomous vehicles, and robots. This research proposes an improved real-time object recognition approach based on the YOLOv9 architecture, trained on a football-specific custom dataset. The 255 annotated images in the dataset are separated into training, validation, and test sets with an 80-15-5 split ratio;there are 204 images in the training set, 38 in the validation set, and 13 in the test set. Our gelan-c model achieves remarkable detection performance with 467 layers, 25,414,044 parameters, and 102.5 GFLOPs. Upon validation, Gelan-c shows an aggregate mAP50 of 75.9%, with mAPs ranging from 32.2% (ball) to 93.7% (goalkeeper) depending on the specified class. The impressive aspect of the Gelan-C is its 31.6 ms picture processing rate, which assures real-time performance. The methodology covers environmental setup, preparation of data, YOLOv9 architecture introduction, and experimental evaluation to demonstrate the model's football-related recognition capabilities. The suggested approach has potential applications in sports event monitoring and evaluation due to its reliability and efficacy in detecting football players, referees, and the ball. © 2024 IEEE.

关键词： Image annotation

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：