检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Yang, Fangli Chang, Liang Qiu, Daowen Pan, Minghua Xiong, Wanjun Guangxi Key Laboratory of Trusted Software Guilin University of Electronic Technology Guilin541004 China Institute of Quantum Computing and Software School of Computer Science and Engineering Sun Yat-sen University Guangzhou510006 China Guangxi Key Laboratory of Cryptography and Information Security Guilin University of Electronic Technology Guilin541004 China

Free-space quantum cryptography has the potential to enable global quantum communication. However, most existing continuous-variable quantum secret sharing (CV-QSS) schemes rely on fiber channels. In this paper, we present a CV-QSS protocol designed for free-space transmission and construct models of crucial parameters, including channel transmittance, excess noise, and interruption probability, thus deriving the bound of the secret key rate. In particular, we provide a multi-source excess noise model for free-space CV-QSS based on the local local oscillator (LLO) scheme and a straightforward optimization of the noise. Furthermore, our research considers the impact of atmospheric turbulence on the protocol. Simulation results demonstrate that the intensification of atmospheric turbulence adversely affects the aforementioned crucial parameters, leading to a significant reduction in the key rate. However, our protocol shows its capability to securely share secrets over a distance of exceeding 60 km among five participants in the presence of turbulence with Cn2 = 1 × 10−15m−2/3, while maintaining a high key rate of approximately 0.55 bit/pulse over a distance of 10 km across twenty participants. These findings suggest that efficient CV-QSS in free space is indeed achievable. This research may serve as a reference for the design and optimization of the practical CV-QSS system. Copyright © 2025, The Authors. All rights reserved.

关键词： Atmospheric turbulence

来源：评论

学校读者我要写书评

暂无评论

DBSSL: A Scheme to Detect Backdoor Attacks in Self-Supervised Learning Models

引用

IEEE Transactions on Dependable and Secure Computing 2025年

作者： Huang, Yuxian Yang, Geng Yuan, Dong Yu, Shui Nanjing University of Posts and Telecommunication College of Computer Science and Software China Jiangsu Key Laboratory of Big Data Security and Intelligent Processing China University of Sydney School of Electrical and Information Engineering Australia University of Technology Sydney School of Computer Science Australia

Recently, self-supervised learning has garnered significant attention for its ability to extract high-quality features from unlabeled data. However, existing research indicates that backdoor attacks can pose significant threats to self-supervised learning. Additionally, due to the substantial training overhead, traditional backdoor detection methods in supervised learning, such as meta-learning, are not well-suited for self-supervised learning. Current backdoor detection methods for self-supervised learning can only defend against backdoor attacks triggered by patch. To address this challenge, we proposes DBSSL, a scheme that can efficiently detect backdoor attacks in self-supervised learning models. In DBSSL, we connect backdoor attacks with adversarial attacks and use adversarial perturbations for detection. Specifically, we first conduct targeted adversarial attacks on the samples, classifying them into different classes. Then, we employ the Z-Score method to analyze these adversarial perturbations. If outliers are present, we can infer that the self-supervised learning model has been backdoored. Unlike previous works, our scheme is better suited for downstream users with constrained computation resources and exhibits excellent detection capabilities for backdoor attacks caused by patch -type or global triggers. Abundant theoretical analysis has demonstrated the feasibility of our scheme. Moreover, extensive experiments show that our method performs well in detecting prevalent backdoor attacks in self-supervised learning, achieving detection accuracy exceeding 95%. © 2004-2012 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Monitoring of Fire-Affected Buildings and Air-Quality Assessment: A Remote-Sensing Study Using Ground-Based interferometric Radar and Google-Earth-Engine

引用

Rangeland Ecology and Management 2025年 101卷 28-42页

作者： Zhang, Hao Zheng, Xiangtian Ghaffar, Bushra Quddoos, Abdul Naz, Iram Aslam, Rana Waqar Fatima, Zoia Abdullah-Al-Wadud, M. Zulqarnain, Rana Muhammad China Academy of Safety Science and Technology Beijing 100012 China Nanjing Institute of Technology School of Computer Engineering Nanjing 211167 China Department of Environmental Science Faculty of Sciences International Islamic University Islamabad 44000 Pakistan State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing (LIESMARS) Wuhan University Wuhan 430079 China Department of Software Engineering College of Computer and Information Sciences King Saud University Riyadh 11543 Saudi Arabia Department of Mathematics Saveetha School of Engineering SIMATS Thandalam Tamilnadu Chennai - 602105 India

This study investigates the environmental and structural impacts of crop and rangeland residue burning in Punjab during April and May 2023, utilizing ground-based interferometric radar alongside atmospheric and land use datasets. The radar system was employed to monitor structural deformations in buildings affected by fire, providing critical insights into the broader impacts of the fires on infrastructure. Multiple ground-based interferometric radars for distributed monitoring can effectively obtain the three-dimensional deformation components of structural targets in real fire scenes. The monitoring results conform to the displacement variation law of existing portal steel frame structures during fire deformation, and can provide effective data support for building structure collapse warning under fire. This approach offers a novel method for assessing the structural integrity of buildings during and after fire events, complementing traditional environmental monitoring. Environmental data show a marked increase in atmospheric pollutants such as NO₂, CH₄, and CO in May 2023, directly linked to the residue burning. Active fire data from FIRMS and MODIS revealed a 98% increase in fire events compared to April, while land surface temperature analysis showed a rise of 1.54%, with affected areas reaching up to 45.9°C. LULC analysis indicated a 23.3% decrease in cropland and rangeland areas, reflecting extensive land clearing. These findings highlight the urgent need for sustainable agricultural management and policies to mitigate the environmental and infrastructural damage caused by residue burning in Punjab. © 2025 The Society for Range Management

关键词： Agriculture Fire & Air Quality Distributed monitoring Google Earth Engine Ground-based interferometric radar Sentinel-2

来源：评论

学校读者我要写书评

暂无评论

CSBNC-PAL: Consistency Semi-supervised Brain Network Classification Framework with Prototypical-Adversarial Learning

引用

IEEE Journal of Biomedical and Health Informatics 2025年 PP卷 PP页

作者： Ji, Junzhong Liu, Gan Wang, Xingyu Beijing University of Technology Beijing Municipal Key Laboratory of Multimedia and Intelligent Software Technology College of Computer Science Beijing Artificial Intelligence Institute Beijing 100124 China

In recent years, semi-supervised learning (SSL) for functional brain network (FBN) classification has gained considerable attention due to its potential to leverage large amounts of unlabeled data from multisite. However, existing SSL methods often struggle to address the distributional differences across different sites, which limits their ability to extract discriminative features from the unlabeled data, thus hindering classification performance. To overcome this challenge, we propose a novel consistency semi-supervised FBN classification framework with prototypical-adversarial learning, termed CSBNC-PAL. Specifically, we first design a contrastive consistency module (CCM) that utilizes contrastive learning to more effectively exploit unlabeled data and learn preliminary feature representations. Then, we introduce a prototype alignment module (PAM) that computes site-aware prototypes through weighted feature clustering to guide inter-site feature alignment, and achieve inter-site equilibrium feature representations. Finally, we develop an adversarial alignment module (AAM) that employs site-discriminative adversarial training based on a gradient reversal layer to guide intra-site feature alignment, and learn site-invariant features. The three modules above are optimized collectively in an end-to-end manner, ensuring effective learning from both labeled and unlabeled data while alleviating the distribution differences of multisite data. Experiments on the ABIDE I, ABIDE II, and ADHD-200 datasets demonstrate that the CSBNC-PAL outperforms many state-of-the-art SSL methods in FBN classification. © 2013 IEEE.

关键词： Brain network classification Feature alignment Multisite fMRI Semi-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Low-shot Video Object Segmentation

引用

IEEE Transactions on Pattern Analysis and Machine Intelligence 2025年 PP卷 PP页

作者： Yan, Kun Wei, Fangyun Dai, Shuyu Wu, Minghui Wang, Ping Xu, Chang Peking University School of Computer Science Beijing100871 China University of Sydney School of Computer Science DarlingtonNSW2008 Australia Peking University School of Software and Microelectronics Beijing100871 China Shanghai Artificial Intelligence Laboratory and Mininglamp Technology China Peking University National Engineering Research Center for Software Engineering Beijing100871 China Peking University School of Software and Microelectronics China China

Prior research in video object segmentation (VOS) predominantly relies on videos with dense annotations. However, obtaining pixel-level annotations is both costly and time-intensive. In this work, we highlight the potential of effectively training a VOS model using remarkably sparse video annotations - specifically, as few as one or two labeled frames per training video, yet maintaining near equivalent performance levels. We introduce this innovative training methodology as low-shot video object segmentation, abbreviated as low-shot VOS. Central to this method is the generation of reliable pseudo labels for unlabeled frames during the training phase, which are then used in tandem with labeled frames to optimize the model. Notably, our strategy is extremely simple and can be incorporated into the vast majority of current VOS models. For the first time, we propose a universal method for training VOS models on one-shot and two-shot VOS datasets. In the two-shot configuration, utilizing just 7.3% and 2.9% of labeled data from the YouTube-VOS and DAVIS benchmarks respectively, our model delivers results on par with those trained on completely labeled datasets. It is also worth noting that in the one-shot setting, a minor performance decrement is observed in comparison to models trained on fully annotated datasets. Code and models are available at https://***/yk-pku/Low-shot-VOS. © 1979-2012 IEEE.

关键词： Video analysis

来源：评论

学校读者我要写书评

暂无评论

ExpertDRL: Request Dispatching and Instance Configuration for Serverless Edge Inference with Foundation Models

引用

IEEE Transactions on Mobile Computing 2025年

作者： Zeng, Yue Zhou, Junlong Ye, Baoliu Qu, Zhihao Guo, Song Gong, Tianjian Li, Pan Nanjing University of Science and Technology Department of Computer Science and Engineering Jiangsu Nanjing210094 China Nanjing University State Key Laboratory for Novel Software Technology Department of Computer Science and Technology Nanjing210023 China Hohai University Key Laboratory of Water Big Data Technology of Ministry of Water Resources Nanjing211100 China Hong Kong University of Science and Technology Hong Kong Southwest University College of Electronic and Information Engineering Chongqing400715 China

The prevalence of the pre-training & fine-tuning paradigm enables machine learning models to quickly adapt to various downstream tasks by fine-tuning pre-trained foundation models (FMs), greatly facilitating various IoT applications that rely on model inference in dynamic edge serverless environments. Efficiently dispatching inference requests and configuring instances to batch inference requests can significantly enhance resource efficiency. However, existing serverless inference solutions are tailored for traditional models, make coarse-grained request dispatching and instance configuration decisions, fail to exploit the shared model backbone characteristics of the FM and capture delayed rewards in dynamic environments, and ignore communication latency between edge sites, resulting in high costs and constraint violations. In this paper, we leverage our insight that fine-grained batch inference requests can effectively exploit the shared model backbone feature of FM to save monetary costs. We propose an algorithm that incorporates deep reinforcement learning (DRL) and expert intervention for fine-grained request dispatching and instance configuration, where the DRL component outputs fractional solutions as guidance, while the expert intervention module integrates our insights—batching reduces monetary costs at the expense of increased inference latency, whereas higher configurations shorten inference latency. This module rounds fractional solutions and adjusts instance configurations to search for optimal solutions while satisfying constraints, with theoretical guarantees rigorously proved. Finally, we conducted our experiments on an OpenFaas-based platform and simulator, and extensive trace-driven evaluation results show that ExpertDRL can save costs by up to 85.14% and improve request acceptance ratio by up to 26.93%, compared to the state-of-the-art solution. © 2002-2012 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

MA-Det: A Discriminative Morphology-Aware Detector for Cervical Lesion Cell Clumps

MA-Det: A Discriminative Morphology-Aware Detector for Cervi...

引用

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

作者： Ziyang Yin Qian Huang Yulin Chen Hao Lu College of Computer Science and Software Engineering Hohai University Nanjing China Key Laboratory of Water Big Data Technology of Ministry of Water Resources Hohai University Nanjing China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Automated detection of cervical lesion cell clumps is crucial for cervical cancer screening. However, the dense packing and overlap of cells, caused by adhesion molecules, make detection challenging. To address this issue, we propose the Morphology-Aware Detector (MA-Det). Specifically, by innovatively employing Deformable Convolution, we propose the Dynamic Context Aggregation Module (DCAM) which dynamically captures contextual information, improving the feature representation of lesion cells. To enhance discrimination in high-density regions, we design the Distance-Weighted Interaction Module (DWIM) and Adaptive Morphology-Aware (AMA) loss into the detection head. These components improve spatial awareness by utilizing interactions between cell features and proposals. In particular, our method achieves state-of-the-art performance on the CDetector and CRIC datasets while significantly reducing parameters.

关键词： Convolution Adhesives Detectors Medical services Feature extraction Complexity theory Lesions Proposals Speech processing Cervical cancer

来源：评论

学校读者我要写书评

暂无评论

Question Selection for Multi-Modal Code Search Synthesis using Probabilistic Version Spaces

引用

IEEE Transactions on software Engineering 2025年

作者： Wu, Jiarong Jiang, Yanyan Wei, Lili Xu, Congying Cheung, Shing-Chi Xu, Chang The Hong Kong University of Science and Technology Department of Computer Science and Engineering Hong Kong McGill University Department of Electrical and Computer Engineering Montreal Canada Nanjing University State Key Laboratory for Novel Software Technology Department of Computer Science and Technology Nanjing China

Searching the occurrences of specific code patterns (code search) is a common task in software engineering, and programming by example (PBE) techniques have been applied to ease customizing code patterns. However, previous PBE tools only synthesize programs meeting the input-output examples, which may not always align with the user intent. To bridge this gap, this paper proposes Excalibur, a multi-modal (example and natural language description) and interactive synthesizer for code search. Excalibur ensures that the generated programs are correct for the provided examples (soundness) and include the user-intended program (bounded completeness). Furthermore, Excalibur helps the user identify the user-intended program through question-answer interaction. To minimize the required interaction efforts, question selection is crucial. To improve question selection for code search, we propose probabilistic version spaces (ProbVS), in which the user-intended program’s probability is high and others are low. ProbVS combines traditional version spaces for compactly representing extensive programs and large language models (on the user-provided natural language description) for adjusting programs’ probabilities to align with users’ intents. Extensive experiments on a benchmark of 44 tasks demonstrated the effectiveness of Excalibur and ProbVS and demystified how ProbVS affects probability distributions and how the configurable parameters affect ProbVS. © 1976-2012 IEEE.

关键词： Normal distribution

来源：评论

学校读者我要写书评

暂无评论

Developers’ Views on Commercial Involvement in OSS - A Survey from Three Projects

引用

IEEE Transactions on software Engineering 2025年

作者： Qin, Mian Zhang, Yuxia Zhou, Minghui Wang, Zhe Li, Haoyang Liu, Hui Beijing Institute of Technology School of Computer Science and Technology Beijing China Peking University School of Computer Science Beijing China Ministry of Education Key Laboratory of High Confidence Software Technologies Beijing China Beihang University School of Humanities and Social Sciences Beijing China OpenInfra Foundation Shanghai China

Given the well-established merits of open source software (OSS), many profit-oriented companies actively participate in OSS communities, making significant contributions. Existing studies have predominantly focused on several advanced and specific questions regarding this phenomenon, mainly from the companies’ perspective, such as companies’ domination and withdrawal. A more basic and comprehensive understanding is missing, i.e., how OSS developers perceive such corporate engagement. Individual developers, including both volunteers and developers assigned by companies, are directly impacted by and have personal experiences with the consequences of commercial participation in OSS projects. This paper aims to bridge this gap by amplifying the voices of individual developers and providing valuable insights that have the potential to enhance companies’ participation in OSS projects. We conducted a survey involving developers from three OSS projects, i.e., Rust, OpenStack, and the Linux kernel, focusing on their attitudes and expectations regarding corporate involvement. We received 84 meaningful responses and analyzed their open-ended responses through thematic analysis. The results suggest that regardless of whether developers were paid or voluntary contributors, a prevailing attitude emerged – 67.9% of developers expressed a positive view of companies’ participation in OSS. The key idea behind their positive attitudes is perceiving commercial participation as a win-win for both the OSS community and companies. The Rust community remains more neutral when compared with the other two communities. We also surveyed and analyzed developers’ expectations of companies’ better participation, which can shed light on how OSS ecosystems can sustainably evolve with the companies involved. © 1976-2012 IEEE.

关键词： Machine oriented languages

来源：评论

学校读者我要写书评

暂无评论

Learned Video Compression With Refined Adaptive Flow Pyramid And Coordinate-Aware Attention

Learned Video Compression With Refined Adaptive Flow Pyramid...

引用

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

作者： Qian Huang Wenting Liu Xin Li Yiming Wang College of Computer Science and Software Engineering Hohai University Nanjing China Key Laboratory of Water Big Data Technology of Ministry of Water Resources Hohai University Nanjing China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

In video compression, motion estimation and motion compensation are critical for achieving efficient encoding. Although the commonly used SpyNet and bilinear interpolation have contributed in improving the compression efficiency, they still have limitations. SpyNet often loses details and fails to fully utilize the feature extraction capabilities of deep networks. Furthermore, bilinear interpolation inherently attenuates high-frequency information, leading to frame blurring and distortion. In this paper, we propose a novel video compression algorithm. To overcome the limitations of SpyNet, we propose a refined adaptive flow pyramid network. This network uses a multi-scale feature pyramid to capture more details. Moreover, we use an iterative cost volume refinement engine that improves the feature representation of the network and iteratively improve the accuracy of motion estimation. In addition, to overcome the limitation of bilinear interpolation, we propose a coordinate-aware attention module, which captures more high-frequency information to improve the accuracy of motion compensation. Experimental results show that our method outperforms VTM-13.2 (LDP) in terms of PSNR.

关键词： Interpolation Adaptive systems Costs Acoustic distortion Accuracy Motion estimation Video compression Feature extraction Motion compensation Speech processing

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：