检索结果-内蒙古大学图书馆

4th International Conference on Educational Technology, ICET 2024

作者： Wang, Hang Chen, Ping Ailiya Shen, Zhiqi Nanyang Technological University College of Computing and Data Science Singapore

ISBN: (纸本)9798350376944

This paper explores the integration of large language models (LLMs) into the curriculum design process for short-term adult training programs, highlighting the role of prompt engineering as a transformative tool. As adult education needs to adapt rapidly to changing professional landscapes, the ability to quickly develop and customize educational content is paramount. This study introduces a framework that employs prompt engineering to leverage LLMs, such as ChatGPT, for generating customized curriculum design. Through a systematic process, curriculum designers input tailored prompts into LLMs to produce relevant and accurate educational content. The methodology involves iterating these prompts based on feedback to refine and enhance the curriculum. A survey conducted among educational professionals, who have varied familiarity with AI, supports the effectiveness of this approach. The results indicate that well-crafted prompts not only expedites the curriculum development process but may also enhance the educational outcomes by providing more customized and adaptive learning experiences. This approach is posited as a promising solution to the challenges faced in adult education, particularly in settings demanding rapid course development and delivery. © 2024 IEEE.

关键词： Curricula

来源：评论

学校读者我要写书评

暂无评论

Adversarial Text Generation with Dynamic Contextual Perturbation

Adversarial Text Generation with Dynamic Contextual Perturba...

引用

2024 IEEE Calcutta Conference, CALCON 2024

作者： Waghela, Hetvi Sen, Jaydip Rakshit, Sneha Dasgupta, Subhasis Praxis Business School Dep of Data Science Kolkata India

ISBN: (纸本)9798331516666

Adversarial attacks on Natural Language Processing (NLP) models expose vulnerabilities by introducing subtle perturbations to input text, often leading to misclassification while maintaining human readability. Existing methods typically focus on word-level or local text segment alterations, overlooking the broader context, which results in detectable or semantically inconsistent perturbations. We propose a novel adversarial text attack scheme named Dynamic Contextual Perturbation (DCP). DCP dynamically generates context-aware perturbations across sentences, paragraphs, and documents, ensuring semantic fidelity and fluency. Leveraging the capabilities of pre-trained language models, DCP iteratively refines perturbations through an adversarial objective function that balances the dual objectives of inducing model misclassification and preserving the naturalness of the text. This comprehensive approach allows DCP to produce more sophisticated and effective adversarial examples that better mimic natural language patterns. Our experimental results, conducted on various NLP models and datasets, demonstrate the efficacy of DCP in challenging the robustness of state-of-the-art NLP systems. By integrating dynamic contextual analysis, DCP significantly enhances the subtlety and impact of adversarial attacks. This study highlights the critical role of context in adversarial attacks and lays the groundwork for creating more robust NLP systems capable of withstanding sophisticated adversarial strategies. © 2024 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

FedDW: Leveraging dataset Distillation for Black-Box Watermarking in Federated Learning Models 8th

FedDW: Leveraging Dataset Distillation for Black-Box Waterm...

引用

8th International Scientific Conference on Intelligent Information Technologies for Industry, IITI 2024

作者： Zhou, Dawei Peng, Jialiang School of Data Science and Technology Heilongjiang University Harbin China

ISBN: (纸本)9783031774102

Federated Learning (FL) is a distributed machine learning paradigm wherein multiple parties collaboratively train a shared model while maintaining the privacy of their respective data. However, this collaborative process grants client’s full access to the global model, thereby exacerbating the risk of model leakage. To mitigate this risk, it is crucial to develop methods for verifying model ownership and tracing malicious clients responsible for model leakage. Existing methods often utilize white-box watermarking techniques, wherein the watermark verification process requires direct access to the model’s internal parameters and structure. This approach is not feasible in many real scenarios, as numerous potentially suspicious models are reluctant to reveal their internal parameters. In this paper, we propose a fully black-box approach that addresses these challenges. Our method involves distilling the original data to establish the foundation of our watermark trigger set. We also propose a feature fusion method to enhance the diversity of the trigger sets, assigning a distinct trigger set to each client. Additionally, we introduce a regularization term to constrain the model parameters before and after watermarking. Experimental results demonstrate that our approach preserves the accuracy of the original task while ensuring ownership verification and tracing leakers. Furthermore, our method shows strong robustness against various watermark removal and detection attacks. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Bi-Jordan n-Derivations on Triangular Rings: Maximal Quotient Rings and Faithful Module

引用

Journal of Mathematical Research with Applications 2024年第5期44卷 596-616页

作者： Xinfeng LIANG Lingling ZHAO School of Mathematics and Big Data Anhui University of Science and Technology

In this paper, we mainly study the structure of bi-Jordan n-derivations in triangular rings under conditions of maximal quotient rings and faithful bimodules, respectively. It is shown that every bi-Jordan n-derivation can be decomposed into the sum of an inner biderivation and an extremal biderivation in two different conditions. As by-products, the structures of bi-Jordan n-derivation over upper triangular matrix rings and nest algebras are characterized, respectively,and generalize the known results.

关键词： triangular rings maximal quotient rings faithful bimodules nest algebras extremal biderivation

来源：评论

学校读者我要写书评

暂无评论

Prediction of Hazardous Asteroids Using Machine Learning 4

Prediction of Hazardous Asteroids Using Machine Learning

引用

4th International Conference on Emerging Systems and Intelligent Computing, ESIC 2024

作者： Surekha, Nemmaluri Purna Nanjundan, Preethi Bashir, Shahid Christ University Pune Department of Data Science Lavasa India

ISBN: (纸本)9798350349856

As the need for early detection and mitigation of potential threats from near-Earth objects continues to grow, this study presents a comprehensive approach to predicting hazardous asteroids through the application of machine learning techniques. With the increasing interest in safeguarding our planet from potential impact events, the accurate classification and prediction of hazardous asteroids is of paramount importance. This research leverages a diverse dataset comprising a wide array of asteroid characteristics, including orbital parameters, physical properties, and historical impact data, to train and validate machine learning models. The study employs a combination of feature engineering, data preprocessing, and state-of-the-art machine learning algorithms to assess the risk posed by asteroids in near-Earth space. © 2024 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

CTSM: Combining Trait and State Emotions for Empathetic Response Model 30

CTSM: Combining Trait and State Emotions for Empathetic Resp...

引用

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

作者： Wang, Yufeng Chen, Chao Yang, Zhou Wang, Shuhui Liao, Xiangwen College of Computer and Data Science Fuzhou University Fuzhou China Digital Fujian Institute of Financial Big Data Fuzhou China China

ISBN: (纸本)9782493814104

Empathetic response generation endeavors to empower dialogue systems to perceive speakers' emotions and generate empathetic responses accordingly. Psychological research demonstrates that emotion, as an essential factor in empathy, encompasses trait emotions, which are static and context-independent, and state emotions, which are dynamic and context-dependent. However, previous studies treat them in isolation, leading to insufficient emotional perception of the context, and subsequently, less effective empathetic expression. To address this problem, we propose Combining Trait and State emotions for Empathetic Response Model (CTSM). Specifically, to sufficiently perceive emotions in dialogue, we first construct and encode trait and state emotion embeddings, and then we further enhance emotional perception capability through an emotion guidance module that guides emotion representation. In addition, we propose a cross-contrastive learning decoder to enhance the model's empathetic expression capability by aligning trait and state emotions between generated responses and contexts. Both automatic and manual evaluation results demonstrate that CTSM outperforms state-of-the-art baselines and can generate more empathetic responses. Our code is available at https://***/wangyufeng-empty/CTSM. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

ANT: Adaptive Noise Schedule for Time Series Diffusion Models 38

ANT: Adaptive Noise Schedule for Time Series Diffusion Model...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Lee, Seunghan Lee, Kibok Park, Taeyoung Department of Statistics and Data Science Yonsei University Korea Republic of

Advances in diffusion models for generative artificial intelligence have recently propagated to the time series (TS) domain, demonstrating state-of-the-art performance on various tasks. However, prior works on TS diffusion models often borrow the framework of existing works proposed in other domains without considering the characteristics of TS data, leading to suboptimal performance. In this work, we propose Adaptive Noise schedule for Time series diffusion models (ANT), which automatically predetermines proper noise schedules for given TS datasets based on their statistics representing non-stationarity. Our intuition is that an optimal noise schedule should satisfy the following desiderata: 1) It linearly reduces the non-stationarity of TS data so that all diffusion steps are equally meaningful, 2) the data is corrupted to the random noise at the final step, and 3) the number of steps is sufficiently large. The proposed method is practical for use in that it eliminates the necessity of finding the optimal noise schedule with a small additional cost to compute the statistics for given datasets, which can be done offline before training. We validate the effectiveness of our method across various tasks, including TS forecasting, refinement, and generation, on datasets from diverse domains. Code is available at this repository: https://***/seunghan96/ANT. © 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

dataset for Identification of Homophobia and Transphobia for Telugu, Kannada, and Gujarati 30

Dataset for Identification of Homophobia and Transphobia for...

引用

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

作者： Kumaresan, Prasanna Kumar Ponnusamy, Rahul Sharma, Dhruv Buitelaar, Paul Chakravarthi, Bharathi Raja Data Science Institute University of Galway Ireland Varanasi India

ISBN: (纸本)9782493814104

Users of social media platforms are negatively affected by the proliferation of hate or abusive content. There has been a rise in homophobic and transphobic content in recent years targeting LGBT+ individuals. The increasing levels of homophobia and transphobia online can make online platforms harmful and threatening for LGBT+ persons, potentially inhibiting equality, diversity, and inclusion. We are introducing a new dataset for three languages, namely Telugu, Kannada, and Gujarati. Additionally, we have created an expert-labeled dataset to automatically identify homophobic and transphobic content within comments collected from YouTube. We provided comprehensive annotation rules to educate annotators in this process. We collected approximately 10,000 comments from YouTube for all three languages. Marking the first dataset of these languages for this task, we also developed a baseline model with pre-trained transformers. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： dataset creation Dravidian Languages Hate speech Homophobia Low-Resourced Languages Transphobia

来源：评论

学校读者我要写书评

暂无评论

Forecasting DDoS attack with machine learning for network forensic investigation

引用

International Journal of Reasoning-based Intelligent Systems 2024年第5期16卷 352-359页

作者： Chatterjee, Saswati Satpathy, Suneeta Paikaray, Bijay Kumar Faculty of Emerging Technologies Sri Sri University Cuttack India Center for Data Science SOA University Odisha India

The recognition of intrusion attempts is the fundamental region of network security, with the objective of identifying the impact of these actions on the distinctive variations of the captured traffic. Innovation and inquiry are now necessary for most of the attacks. In this paper, a machine-learning approach has been used to track anomalous network traffic. Also, the statistical parameters are employed to enhance the performance based on learning models. The paper also represents a detection pattern through machine learning experiments for DDoS attacks on KDD Cup 99 dataset where ten features are used for accuracy measures. The study further classifies the hidden forms to sense the DDoS attack patterns. The paper further concludes with the experimental outcomes that establish the improved performance assessment of the K-nearest neighbour algorithm in comparison to other predictable learning approaches and thus the proposed model can reach the uppermost correctness. Copyright © 2024 Inderscience Enterprises Ltd.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

MODIFIED BRASCAMP-LIEB INEQUALITIES AND LOG-SOBOLEV INEQUALITIES FOR ONE-DIMENSIONAL LOG-CONCAVE MEASURE

引用

Acta Mathematica Scientia 2025年第1期45卷 104-117页

作者： Denghui WU Jiazu ZHOU College of Science Northwest A&F UniversityYangling712100China School of Mathematics and Big Data Guizhou Education UniversityGuiyang550018China

In this paper,we develop Maurey’s and Bobkov-Ledoux’s methods to prove modified Brascamp-Lieb inequalities and log-Sobolev inequalities for one-dimensional log-concave *** prove these inequalities,the harmonic Prékopa-Leindler inequality is *** prove that these new inequalities are more efficient in estimating the variance and entropy for some functions with exponential terms.

关键词： Brunn-Minkowski inequality Prékopa-Leindler inequality Brascamp-Lieb inequality log-Sobolev inequality log-concave measure

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：