检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Rütte, Dimitri Von Fluri, Janis Ding, Yuhui Orvieto, Antonio Schölkopf, Bernhard Hofmann, Thomas Data Analytics Lab Department of Computer Science ETH Zurich Switzerland ELLIS Institute Tübingen Tübingen AI Center Germany Max Planck Institute for Intelligent Systems Tübingen Germany

While state-of-the-art language models achieve impressive results through next-token prediction, they have inherent limitations such as the inability to revise already generated tokens. This has prompted exploration of alternative approaches such as discrete diffusion. However, masked diffusion, which has emerged as a popular choice due to its simplicity and effectiveness, reintroduces this inability to revise words. To overcome this, we generalize masked diffusion and derive the theoretical backbone of a family of general interpolating discrete diffusion (GIDD) processes offering greater flexibility in the design of the noising processes. Leveraging a novel diffusion ELBO, we achieve compute-matched state-of-the-art performance in diffusion language modeling. Exploiting GIDD’s flexibility, we explore a hybrid approach combining masking and uniform noise, leading to improved sample quality and unlocking the ability for the model to correct its own mistakes, an area where autoregressive models notoriously have struggled. Our code and models are open-source: https://***/dvruette/gidd/ © 2025, CC BY.

关键词： Modeling languages

来源：评论

学校读者我要写书评

暂无评论

One-Dimensional EEG Artifact Removal Network Based on Convolutional Neural Networks

Journal of Network Intelligence

引用

Journal of Network Intelligence 2024年第1期9卷 142-159页

作者： Xiong, Jun Meng, Xiang-Long Chen, Zhao-Qi Wang, Chuan-Sheng Zhang, Fu-Quan Grau, Antoni Chen, Yang Huang, Jing-Wei School of Computer and Data Science Minjiang University Fuzhou University Town No. 200 Xiyuangong Road Fuzhou China College of Electronic Engineering Shandong University of Science and Technology No. 579 Qianwangang Road Huangdao District Qingdao China College of Computer and Big Data Fuzhou University Fuzhou University Town No. 2 Wulongjiang North Road Fuzhou China Department of Automatic Control Technical Polytechnic University of Catalonia Autonomous Region of Catalonia Barcelona Spain Digital Media Art Key Laboratory of Sichuan Province Sichuan Conservatory of Music Fuzhou Technology Innovation Center of intelligent Manufacturing information System Minjiang University Fuzhou University Town No. 200 Xiyuangong Road Fuzhou China Fujian Province University No. 1 Campus New Village Longjiang Street Fuqing China School of Mechanical and Automotive Engineering Fujian University of Technology No. 33 Xuefu South Road University New District Fuzhou China

The electroencephalogram (EEG) serves as a significant tool in the realms of clinical medicine, cerebral investigation, and neurological disorders research. However, the EEG records we obtain are often easily contaminated by various artifacts, which can blur or distort the underlying EEG signals and make data interpretation difficult. Generally speaking, removing EEG artifacts is considered an essential step in brain signal analysis. Therefore, removing artifacts is crucial for obtaining accurate and reliable EEG signals for subsequent analysis. Recently, deep learning techniques have found widespread application across various domains for denoising tasks, including image denoising and EEG denoising. Many advanced algorithms have been developed in image denoising, which has achieved good results in enhancing low-quality images. Moreover, it has shown superior performance in EEG denoising. In contrast, few people have devoted themselves to studying EEG denoising, and existing convolutional neural network EEG denoising methods still have problems of overfitting and poor denoising effect in Electromyograph(EMG) and ElectroOculoGram(EOG) artifact removal. Therefore, this paper proposes a method called DWINet (De-artifacting with Image-based Network for EEG Signals) based on an image dehazing network DRHNet for removing artifacts from EEG signals. Specifically, our approach DWINet, addresses the de-artifacting issue in EEG signals by converting it as an image dehazing problem and utilizes the image dehazing capability of DRHNet to enhance the denoising performance of EEG signals. Experimental results demonstrate that the proposed method outperforms the compared algorithms in removing the ocular artifact in EEG signals and exhibits higher accuracy and robustness. © 2024, J. Network Intell. All rights reserved.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of Gaze Estimation Models

A Comparative Study of Gaze Estimation Models

引用

Information Technology (ACIT)

作者： Abdallah Moubayed MohammadNoor Injadat Mohammad Kanan Computer Engineering Department Interdisciplinary Research Center for Intelligent Secure Systems King Fahd University of Petroleum & Minerals Dhahran Saudi Arabia Data Science & Artificial Intelligence Department Zarqa University Zarqa Jordan Department of Industrial Engineering University of Business and Technology Jeddah Saudi Arabia

ISBN: (数字)9798331540012

ISBN: (纸本)9798331540029

The eye-mind hypothesis suggests that people tend to look at what they’re actively thinking about, forming the basis of eye and gaze tracking. This concept is gaining attention in deep learning due to its broad applications. The use of AI alongside webcams to monitor eye movements is increasingly popular and expected to grow further. This is further emphasized by recent data showing a growing use of eye gaze estimation techniques, especially in marketing research, e-commerce, and educational tools. Accordingly, multiple previous research works have developed various eye and gaze estimation and tracking models. However, one main limitation is that many models use their own datasets for performance evaluation as well as having different underlying computing resources that are used during training. Consequently, it becomes harder to compare the effectiveness and efficiency of these models. To that end, this work aims at providing a comprehensive comparative study of three well-established eye gaze estimation models, namely OpenGaze, GazeRefineNet, ODABE, and FAZE models using a unified evaluation framework. Experimental results conducted using GazeCapture dataset illustrate that OpenGaze model achieves a mean error of 2.27 cm mean error, GazeRefineNet model achieves 1.91 cm, ODABE model achieves 3.46 cm, and FAZE model achieves 2.9 cm. This indicates that GazeRefineNet outperforms the other models in terms of accuracy while having comparable computational complexity.

关键词： Training Performance evaluation Webcams Tracking Computational modeling Estimation Gaze tracking Electronic commerce Information technology Monitoring

来源：评论

学校读者我要写书评

暂无评论

IDENTIFYING SOURCE SPEAKERS FOR VOICE CONVERSION BASED SPOOFING ATTACKS ON SPEAKER VERIFICATION SYSTEMS

arXiv

引用

arXiv 2022年

作者： Cai, Danwei Cai, Zexin Li, Ming Department of Electrical and Computer Engineering Duke University Durham United States Data Science Research Center Duke Kunshan University Kunshan China

An automatic speaker verification system aims to verify the speaker identity of a speech signal. However, a voice conversion system could manipulate a person's speech signal to make it sound like another speaker's voice and deceive the speaker verification system. Most countermeasures for voice conversion-based spoofing attacks are designed to discriminate bona fide speech from spoofed speech for speaker verification systems. In this paper, we investigate the problem of source speaker identification - inferring the identity of the source speaker given the voice converted speech. To perform source speaker identification, we simply add voice-converted speech data with the label of source speaker identity to the genuine speech dataset during speaker embedding network training. Experimental results show the feasibility of source speaker identification when training and testing with converted speeches from the same voice conversion model(s). In addition, our results demonstrate that having more converted utterances from various voice conversion model for training helps improve the source speaker identification performance on converted utterances from unseen voice conversion models. © 2022, CC BY.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

Active Strategy for Learning Non-Deterministic Automata by Peer

Active Strategy for Learning Non-Deterministic Automata by P...

引用

Business Analytics for Technology and Security (ICBATS), International Conference on

作者： Bisma Shahid Abd Ur Rehman Nafisa Tahir Omar Sattar Department of Computer Science Riphah International Uuniversity Lahore Pakistan School of Computer Science NCBA&E Lahore Pakistan Applied Science Research Center Applied Science Private University Amman Jordan General Education Department Skyline University College University City Sharjah Sharjah UAE

Finite Automata plays a vital role as a course in computer science. This subject is so much challenging and tough work for the students because they found this course very less attractive and not able to understand it in easy way. And this course prerequisite subject is mathematics. In this paper we present the active strategy in peer group. There are different kinds of activities we apply on the group so they find out the course attractive and easy to learn with the help of their peers. Due to which they able to learn Non-Deterministic Finite Automata and they also use simulation software for learning procedure to improve it.

关键词： computer science Learning automata Automata Software Mathematics Security Business

来源：评论

学校读者我要写书评

暂无评论

ECLIPSE: Expunging Clean-label Indiscriminate Poisons via Sparse Diffusion Purification

arXiv

引用

arXiv 2024年

作者： Wang, Xianlong Hu, Shengshan Zhang, Yechao Zhou, Ziqi Zhang, Leo Yu Xu, Peng Wan, Wei Jin, Hai National Engineering Research Center for Big Data Technology and System China Services Computing Technology and System Lab China Cluster and Grid Computing Lab China Hubei Engineering Research Center on Big Data Security China Hubei Key Laboratory of Distributed System Security China School of Cyber Science and Engineering Huazhong University of Science and Technology Wuhan430074 China School of Computer Science and Technology Huazhong University of Science and Technology Wuhan430074 China School of Information and Communication Technology Griffith University SouthportQLD4215 Australia

Clean-label indiscriminate poisoning attacks add invisible perturbations to correctly labeled training images, thus dramatically reducing the generalization capability of the victim models. Recently, defense mechanisms such as adversarial training, image transformation techniques, and image purification have been proposed. However, these schemes are either susceptible to adaptive attacks, built on unrealistic assumptions, or only effective against specific poison types, limiting their universal applicability. In this research, we propose a more universally effective, practical, and robust defense scheme called ECLIPSE. We first investigate the impact of Gaussian noise on the poisons and theoretically prove that any kind of poison will be largely assimilated when imposing sufficient random noise. In light of this, we assume the victim has access to an extremely limited number of clean images (a more practical scene) and subsequently enlarge this sparse set for training a denoising probabilistic model (a universal denoising tool). We then introduce Gaussian noise to absorb the poisons and apply the model for denoising, resulting in a roughly purified dataset. Finally, to address the trade-off of the inconsistency in the assimilation sensitivity of different poisons by Gaussian noise, we propose a lightweight corruption compensation module to effectively eliminate residual poisons, providing a more universal defense approach. Extensive experiments demonstrate that our defense approach outperforms 10 state-of-the-art defenses. We also propose an adaptive attack against ECLIPSE and verify the robustness of our defense scheme. Our code is available at https://***/CGCL-codes/ECLIPSE. Copyright © 2024, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Correlation Analysis and Predictive Factors for Building a Mathematical Model 1

引用

7th Computational Methods in Systems and Software, CoMeSySo 2023

作者： Nelyub, V.A. Tynchenko, V.S. Gantimurov, A.P. Degtyareva, K.V. Kukartseva, O.I. Artificial Intelligence Technology Scientific and Education Center Bauman Moscow State Technical University Moscow105005 Russia Peter the Great St. Petersburg Polytechnic University Saint Petersburg Russia Information-Control Systems Department Institute of Computer Science and Telecommunications Reshetnev Siberian State University of Science and Technology Krasnoyarsk660037 Russia Department of Technological Machines and Equipment of Oil and Gas Complex School of Petroleum and Natural Gas Engineering Siberian Federal University Krasnoyarsk660041 Russia Department of Information Economic System Institute of Engineering and Economics Reshetnev Siberian State University of Science and Technology Krasnoyarsk660037 Russia Department of Systems Analysis and Operations Research Institute of Informatics and Telecommunications Reshetnev Siberian State University of Science and Technology Krasnoyarsk660037 Russia

ISBN: (数字)9783031535499

ISBN: (纸本)9783031535482

The study, published in the journal Nature Medicine, looked at data on 1,000 people from China who were tracked over an average period of six years. The participants were divided into two groups: those who lived in areas with high levels of air pollution and those who lived in areas with low levels of air pollution. The study analyzed data on patients with lung cancer, including their age, gender, exposure to air pollution, alcohol consumption, dust allergy, occupational hazards, genetic risk, chronic lung disease, balanced diet, obesity, smoking, passive smoking, chest pain, cough, hemoptysis, fatigue, weight loss, shortness of breath, wheezing, difficulty swallowing, nail thickening and snoring. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Air pollution

来源：评论

学校读者我要写书评

暂无评论

Facial Action Unit Recognition with Micro-Action-Aware Transformer 2nd

Facial Action Unit Recognition with Micro-Action-Aware Tran...

引用

2nd CSIG Conference on Emotional Intelligence, CEI 2024

作者： Yuan, Yichen Cheng, Yifan Shao, Zhiwen Dang, Qianwen Chen, Rui Fu, Mingjian Jiang, Shengtian Li, Chunyu Ma, Lizhuang School of Computer Science and Technology China University of Mining and Technology Xuzhou China Mine Digitization Engineering Research Center of the Ministry of Education Xuzhou China Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai China

ISBN: (纸本)9789819650835

Facial action unit (AU) recognition is a challenging task, due to the subtlety of each AU and the correlations among AUs in global face. However, the learning of local-global features has not been thoroughly exploited in most of the existing methods. In this paper, we propose a novel micro-action-aware transformer to integrate local and global feature extractions, which effectively captures subtle AU details while maintaining the global relational modeling capacity of transformers. Besides, we jointly train facial AU recognition and facial landmark detection, in which the two correlated tasks contribute to each other and further facilitate the learning of local-global AU-related feature. Extensive experiments demonstrate that our approach achieves comparable performance to the state-of-the-art AU recognition methods on the challenging BP4D and GFT benchmarks, and works well for landmark detection. Particularly, our approach achieves average F1 score results of 63.3% and 55.8% on BP4D and GFT datasets, respectively. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： facial action unit recognition facial landmark detection Micro-action-aware transformer

来源：评论

学校读者我要写书评

暂无评论

Beyond normal: on the evaluation of mutual information estimators 23

Beyond normal: on the evaluation of mutual information estim...

引用

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Paweł Czyż Frederic Grabowski Julia E. Vogt Niko Beerenwinkel Alexander Marx Department of Biosystems Science and Engineering ETH Zurich and ETH AI Center ETH Zurich Institute of Fundamental Technological Research Polish Academy of Sciences Department of Computer Science ETH Zurich and SIB Swiss Institute of Bioinformatics Department of Biosystems Science and Engineering ETH Zurich and SIB Swiss Institute of Bioinformatics ETH AI Center ETH Zurich and Department of Computer Science ETH Zurich

Mutual information is a general statistical dependency measure which has found applications in representation learning, causality, domain generalization and computational biology. However, mutual information estimators are typically evaluated on simple families of probability distributions, namely multivariate normal distribution and selected distributions with one-dimensional random variables. In this paper, we show how to construct a diverse family of distributions with known ground-truth mutual information and propose a language-independent benchmarking platform for mutual information estimators. We discuss the general applicability and limitations of classical and neural estimators in settings involving high dimensions, sparse interactions, long-tailed distributions, and high mutual information. Finally, we provide guidelines for practitioners on how to select appropriate estimator adapted to the difficulty of problem considered and issues one needs to consider when applying an estimator to a new data set.

关键词：

来源：评论

学校读者我要写书评

暂无评论

The DKU-DukeECE Diarization System for the VoxCeleb Speaker Recognition Challenge 2022

arXiv

引用

arXiv 2022年

作者： Wang, Weiqing Qin, Xiaoyi Cheng, Ming Zhang, Yucong Wang, Kangyue Li, Ming Data Science Research Center Duke Kunshan University Kunshan China Department of Electrical and Computer Engineering Duke University Durham United States

This paper discribes the DKU-DukeECE submission to the 4th track of the VoxCeleb Speaker Recognition Challenge 2022 (VoxSRC-22). Our system contains a fused voice activity detection model, a clustering-based diarization model, and a target-speaker voice activity detection-based overlap detection model. Overall, the submitted system is similar to our previous year's system in VoxSRC-21. The difference is that we use a much better speaker embedding and a fused voice activity detection, which significantly improves the performance. Finally, we fuse 4 different systems using DOVER-lap and achieve 4.75% of the diarization error rate, which ranks the 1st place in track 4. Copyright © 2022, The Authors. All rights reserved.

关键词： Speech recognition

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：