检索结果-内蒙古大学图书馆

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Zhang, Yujun Chen, Yanqu Wang, Jiakai Hu, Jin Tao, Renshuai Liu, Xianglong State Key Laboratory of Complex & Critical Software Environment Beihang University China School of Computer Science and Engineering Beihang University China College of Computer Science Beijing University of Technology China Zhongguancun Laboratory China School of Computer and Information Technology Beijing Jiaotong University China Institute of Data Space Hefei Comprehensive National Science Center China

ISBN: (纸本)9798350368741

There is a growing concern about adversarial attacks against automatic speech recognition (ASR) systems. Although research into targeted universal adversarial examples (AEs) has progressed, current methods are constrained by inefficient exploitation of audio features, demonstrating insufficient attack ability and robustness in the physical world. To solve this problem, we propose a phoneme-tailored attack (PTA) to improve the quality of the generated AEs. Specifically, to improve attack ability, we propose a Diverse Audio Composition Enrichment method, which enhances the utilization of audio features through phoneme-level slicing and recombination. To adapt AEs to complex environments, we propose a Natural Noise Pattern Guidance method to align AEs with natural noise patterns to improve their robustness. Experiments show that our method achieves an average accuracy of more than 72.34% and 98% with and without a norm constraint, and also demonstrates excellent performance in terms of generalization across datasets and resilience to MP3 compression. © 2025 IEEE.

关键词： Automatic speech recognition natural noise pattern phoneme-level targeted universal adversarial examples

来源：评论

学校读者我要写书评

暂无评论

A Practical Learning-Based FTL for Memory Constrained Mobile Flash Storage

A Practical Learning-Based FTL for Memory Constrained Mobile...

引用

2025 Design, Automation and Test in Europe Conference, DATE 2025

作者： Du, Zelin Huang, Kecheng Wang, Tianyu Yao, Xin Chen, Renhai Shao, Zili The Chinese University of Hong Kong Department of Computer Science and Engineering Hong Kong College of Computer Science and Software Engineering Shenzhen University Shenzhen China Huawei Technologies China

ISBN: (纸本)9783982674100

The rapidly growing mobile market is pushing flash storage manufacturers to expand capacity into the terabyte range. However, this presents a significant challenge for mobile storage management: more logical-to-physical page mappings are desired to be efficiently managed and cached while the available caching space is extremely limited. This motivates us to shift toward a new learning-based paradigm: rather than maintaining mappings for individual pages, the learning-based approach can represent mapping relationships for a set of continuous pages. However, to construct linear models, existing methods that either consume the already-limited memory space or reuse flash garbage collection demonstrate poor model construction capabilities or significantly degrade flash performance, making them impractical for real-world use. In this paper, we propose LFTL, a practical, learning-based on-demand flash translation layer design for flash management in mobile devices. In contrast to prior work that centered around gathering sufficient mappings for linear model construction, our key insight is that linear patterns can be extracted and refined by leveraging the orderly, LPA-aligned write stream typical of mobile devices. By doing this, highly accurate linear models can be constructed regardless of the constraints of mobile device's cache limitation. We have implemented a fully functional prototype of LFTL based on FEMU. Our evaluation results show that LFTL is more adaptable to memory-constrained storage devices than state-of-the-art learning-based approaches. © 2025 EDAA.

关键词： flash translation layer learned index Mobile flash storage

来源：评论

学校读者我要写书评

暂无评论

MVDC : A Multi-view Dental Completion Model Based on Contrastive Learning

MVDC : A Multi-view Dental Completion Model Based on Contras...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Xunyu Yang Qingxin Deng Minghan Huang Landu Jiang Dian Zhang College of Computer Science and Software Engineering Shenzhen University Guangdong China College of Future Technology HKUST(GZ) Guangdong China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Restoring the patient’s occlusal function of broken teeth is a challenging task since tooth texture is very complex, a slight deviation may affect the patient’s chewing function and temporomandibular joint function. Therefore, how to efficiently repair the complete shape and real surface of the crown is a critical problem. Traditional technologies are hard to restore complete shape of the dental crown or lack inlay surface details, due to dataset limitations and complexity of missing parts. In this paper, we propose a multi-view crown restoration framework MVDC based on contrastive learning. Specifically, MVDC contains: 1) a multi-view generator with a specially designed loss measurement by using contrastive learning; 2) a multi-scale discriminator mechanism able to consider relation and consistency between teeth from different scales; 3) an occlusal groove extraction network to extract the occlusal details. We conducted extensive experiments on existing public datasets. The results showcase the superior performance of MVDC.

关键词： Correlation Shape Teeth Contrastive learning Signal processing Maintenance engineering Loss measurement Dentistry Surface texture Speech processing

来源：评论

学校读者我要写书评

暂无评论

Generative Positive-Unlabeled Classification for Hunting Small Open Reading Frames

Generative Positive-Unlabeled Classification for Hunting Sma...

引用

2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

作者： Chen, Jie Liao, Wenbin Wen, Du Li, Jianqiang Wang, Fangzhong Shenzhen University College of Computer Science and Software Engineering China Shenzhen University National Engineering Laboratory for Big Data System Computing Technology China Tianjin University Laboratory of Synthetic Microbiology School of Chemical Engineering & Technology China

ISBN: (纸本)9798350337488

The annotation of Open Reading Frames (ORFs) is a crucial step in gene annotation, as it precisely delineates the specific regions of expressed genes. However, small Open Reading Frames (smORFs), in comparison to ORFs, are shorter in length, exhibit lower expression abundance, and are more challenging to predict. Particularly in the presence of noise in prokaryotic data and limited availability of positive sample data, the difficulty of prediction is amplified. Therefore, it is necessary to study smORF prediction methods. However, current machine learning models use limited data for modeling and overlook the existence of undiscovered positive samples within the negative samples. Additionally, they do not incorporate prior knowledge that can be calibrated to enhance the 3-nt periodicity. This work utilizes a multimodal VAE for data dimensionality reduction and employs a GAN to generate latent vectors for data augmentation. It incorporates PU learning to leverage unknown samples and combines Riboseq data from experiments with and without antibiotic treatment. Additionally, an adversarial training mechanism is employed to enhance the model's robustness. © 2023 IEEE.

关键词： Class Imbalance Data Generation Gene Annotation Positive-Unlabeled Learning

来源：评论

学校读者我要写书评

暂无评论

Advanced QUasi-Affine TRansformation Evolutionary (QUATRE) Algorithm and Its Application for Neural Network 4th

Advanced QUasi-Affine TRansformation Evolutionary (QUATRE) A...

引用

4th International Conference on Smart Vehicular Technology, Transportation, Communication and Applications, VTCA 2021

作者： Hu, Pei Pan, Jeng-Shyang Chu, Shu-Chuan College of Computer Science and Engineering Shandong University of Science and Technology Qingdao266590 China School of Computer and Software Nanyang Institute of Technology Nanyang473004 China College of Science and Engineering Flinders University Sturt Road Bedford ParkSA5042 Australia

ISBN: (纸本)9789811640384

The QUasi-Affine TRansformation Evolution (QUATRE) was first proposed by Meng et al. in 2016. It has the characteristics of few parameters and fast convergence. This paper brings two methods to improve its solution quality. The opposite position and comprehensive learning greatly advance the ability of jumping out of local traps when the QUATRE falls into stagnation. Their performance is verified by 23 benchmark functions. In the end, they succeed to train the parameters of neural network and predict the long-term traffic flow in Qingdao. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

ReForm-Eval: Evaluating Large Vision Language Models via Unified Re-Formulation of Task-Oriented Benchmarks 24

ReForm-Eval: Evaluating Large Vision Language Models via Uni...

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Li, Zejun Wang, Ye Du, Mengfei Liu, Qingwen Wu, Binhao Zhang, Jiwen Zhou, Chengxing Fan, Zhihao Fu, Jie Chen, Jingjing Wei, Zhongyu Huang, Xuanjing School of Data Science Fudan University Shanghai China Academy for Engineering and Technology Fudan University Shanghai China Institute of Science and Technology for Brain-Inspired Intelligence Fudan University Shanghai China Software College Northeastern University Liaoning Shenyang China Alibaba Group Zhejiang Hangzhou China Department of Computer Science and Engineering Hong Kong University of Science and Technology Hong Kong School of Computer Science Fudan University Shanghai China

ISBN: (纸本)9798400706868

Recent years have witnessed remarkable progress in the development of large vision-language models (LVLMs). Benefiting from the strong language backbones and efficient cross-modal alignment strategies, LVLMs exhibit surprising capabilities to perceive visual signals and perform visually grounded reasoning. However, the capabilities of LVLMs have not been comprehensively and quantitatively evaluated. Most existing multi-modal benchmarks require task-oriented input-output formats, posing great challenges to automatically assess the free-form text output of LVLMs. To effectively leverage the annotations available and reduce the manual efforts required for constructing new benchmarks, we propose to re-formulate existing benchmarks into unified LVLM-compatible formats. Through systematic data collection and reformulation, we present ReForm-Eval benchmark, offering substantial data for evaluating various capabilities of LVLMs. Through extensive experiments and analysis in ReForm-Eval, we demonstrate the comprehensiveness and reliability of ReForm-Eval in assessing various LVLMs. Our benchmark and evaluation framework is now available at https://***/FudanDISC/ReForm-Eval © 2024 ACM.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

OSCO: An Efficient Segment Routing Scheme for Backup Path 19

OSCO: An Efficient Segment Routing Scheme for Backup Path

引用

2022 IEEE SmartWorld, 19th IEEE International Conference on Ubiquitous Intelligence and Computing, 2022 IEEE International Conference on Autonomous and Trusted Vehicles Conference, 22nd IEEE International Conference on Scalable Computing and Communications, 2022 IEEE International Conference on Digital Twin, 8th IEEE International Conference on Privacy Computing and 2022 IEEE International Conference on Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PriComp/Metaverse 2022

作者： Yang, Shu Li, Jiaming Cui, Laizhong Shenzhen University College of Computer Science and Software Engineering Shenzhen518060 China Peng Cheng Laboratory Shenzhen518060 China

ISBN: (纸本)9798350346558

As artificial intelligence enters the stage of ubiquitous intelligence, the explosive growth of data makes some delay-sensitive services face greater challenges. The environment of ubiquitous intelligence deployment is uncertain, and link failures often occur. Fast Reroute(FRR) can restore communication through backup paths in time after a link failure, but it encounters the loop problem of forwarding in traditional network forwarding. Segment Routing can avoid loops and forward efficiently under the decision of the controller. But the backup path deployed by the controller will bring additional control overhead. The redundant part of the backup path makes the network bear unnecessary load. So we propose an optimization scheme that greatly reduces the control overhead of the backup path with Segment Routing(OSCO). OSCO only labels the hot points along the backup paths and calculates the backup path with the goal of low control overhead. We use three algorithms to compute the backup path and deploy it to the real networks and generated topologies to evaluate the performance. The experimental results show that OSCO can reduce the control overhead of the backup path by up to 99%. © 2022 IEEE.

关键词： Controllers

来源：评论

学校读者我要写书评

暂无评论

Collaborative Streaming and Super Resolution Adaptation for Mobile Immersive Videos 42

Collaborative Streaming and Super Resolution Adaptation for ...

引用

42nd IEEE International Conference on computer Communications, INFOCOM 2023

作者： Zhang, Lei Guo, Haotian Dong, Yanjie Wang, Fangxin Cui, Laizhong Leung, Victor C. M. Shenzhen University College of Computer Science and Software Engineering Shenzhen China The Chinese University of HongKong Sse and FNii Shenzhen China Peng Cheng Laboratory Shenzhen China Shenzhen China The University of British Columbia Department of Electrical and Computer Engineering Vancouver Canada

ISBN: (纸本)9798350334142

Tile-based streaming and super resolution are two representative technologies adopted to improve bandwidth efficiency of immersive video steaming. The former allows selective download of contents in the user viewport by splitting the video into multiple independently decodable tiles. The latter leverages client-side computation to reconstruct the received video into higher quality using advanced neural network models. In this work, we propose CASE, a collaborated adaptive streaming and enhancement framework for mobile immersive videos, which integrates super resolution with tile-based streaming to optimize user experience with dynamic bandwidth and limited computing capability. To coordinate the video transmission and reconstruction in CASE, we identify and address several key design issues including unified video quality assessment, computation complexity model for super resolution, and buffer analysis considering the interplay between transmission and reconstruction. We further formulate the quality-of-experience (QoE) maximization problem for mobile immersive video streaming and propose a rate adaptation algorithm to make the best decisions for download and for reconstruction based on the Lyapunov optimization theory. Extensive evaluation results validate the superiority of our proposed approach, which presents stable performance with considerable QoE improvement, while enabling trade-off between playback smoothness and video quality. © 2023 IEEE.

关键词： Video streaming

来源：评论

学校读者我要写书评

暂无评论

Public integrity verification for data sharing in cloud with asynchronous revocation

引用

Digital Communications and Networks 2022年第1期8卷 33-43页

作者： Wei Song Yu Wu Yihui Cui Qilie Liu Yuan Shen Zicheng Qiu Jianjun Yao Zhiyong Peng School of Computer Science Wuhan UniversityWuhan430074China b School of Software Engineering Chongqing University of Posts and TelecommunicationsChongqing400054China School of Communication and Information Engineering Chongqing University of Posts and TelecommunicationsChongqing400054China College of Information Engineering Tarim UniversityAlar CityXinjiang843300China

Cloud data sharing service,which allows a group of people to access and modify the shared data,is one of the most popular and efficient working styles in ***,there is an uprising trend that enterprises tend to move their IT service from local to cloud to ease the management and reduce the *** the new cloud environment,the cloud users require the data integrity verification to inspect the data service at the cloud *** recent studies have focused on this application *** these studies,each user within a group is required to sign a data block created or modified by *** a user is revoked,all the data previously signed by him should be *** the existing research,the resigning process is dependent on the revoked ***,cloud users are *** may exit the system at any time without notifying the system admin and even are revoked due to *** the developers in the cloud-based software development platform,they are voluntary and not strictly controlled by the *** to this feature,cloud users may not always follow the cloud service *** may not participate in generating the resigning key and may even expose their secret keys after being *** the signature is not resigned in time,the subsequent verification will be *** if the secret key is exposed,the shared data will be maliciously modified by the attacker who grasps the ***,forcing a revoked user to participate in the revocation process will lead to efficiency and security *** a result,designing a practical and efficient integrity verification scheme that supports this scenario is highly *** this paper,we identify this challenging problem as the asynchronous revocation,in which the revocation operations(i.e.,re-signing key generation and resigning process)and the user's revocation are *** the revocation operations must be able to be performed without the participation of the revo

关键词： Cloud storage Integrity verification Asynchronous revocation

来源：评论

学校读者我要写书评

暂无评论

Personalized Alarming System for Asthma Management Based on Lung Functionality: Observational Study

Personalized Alarming System for Asthma Management Based on ...

引用

2023 Advances in science and engineering Technology International Conferences, ASET 2023

作者： Alblooshi, Fatima Albadi, Reem Alblooshi, Khawla Ishimirwe, Emma Uwisoneye, Yvettte Alnajjar, Fady Mehiar, Dua'a College of Information Technology United Arab Emirates Universty Al Ain United Arab Emirates College of It United Arab Emirates Universty Department of Computer Science and Software Engineering Al Ain United Arab Emirates

ISBN: (纸本)9781665454742

Asthma is a chronic respiratory disease that can be challenging to manage. To improve asthma management, we designed a personalized alarm system that utilizes lung function data to monitor and alert individuals to potentially harmful air quality. Our pilot observation included seven individuals who were exposed to two different air-quality environments, and we observed significant variations in their lung function responses. Based on these observations, we developed customized alarm systems for each participant that continuously monitored air quality and provided notifications when it reached levels that could potentially exacerbate their asthma symptoms. This personalized approach has the potential to enable individuals to proactively protect their respiratory health and avoid asthma attacks by taking timely and appropriate action based on their own specific needs. © 2023 IEEE.

关键词： Alarm systems

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：