检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Chen, Huiqiang Zhu, Tianqing Wang, Linlin Yu, Xin Gao, Longxiang Zhou, Wanlei Faculty of Data Science City University of Macau China School of Computer Science University of Queensland QLD Australia Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology Shandong Academy of Sciences Jinan China Shandong Provincial Key Laboratory of Computing Power Internet and Service Computing Shandong Fundamental Research Center for Computer Science Jinan China

Large-scale text-to-image (T2I) diffusion models have revolutionized image generation, enabling the synthesis of highly detailed visuals from textual descriptions. However, these models may inadvertently generate inappropriate content, such as copyrighted works or offensive images. While existing methods attempt to eliminate specific unwanted concepts, they often fail to ensure complete removal—allowing the concept to reappear in subtle forms. For instance, a model may successfully avoid generating images in Van Gogh’s style when explicitly prompted with "Van Gogh", yet still reproduce his signature artwork when given the prompt "Starry Night". In this paper, we propose SAFER, a novel and efficient approach for thoroughly removing target concepts from diffusion models. At a high level, SAFER is inspired by the observed low-dimensional structure of the text embedding space. The method first identifies a concept-specific subspace Sc associated with the target concept c. It then projects the prompt embeddings onto the complementary subspace of Sc, effectively erasing the concept from the generated images. Since concepts can be abstract and difficult to fully capture using natural language alone, we employ textual inversion to learn an optimized embedding of the target concept from a reference image. This enables more precise subspace estimation and enhances removal performance. Furthermore, we introduce a subspace expansion strategy to ensure comprehensive and robust concept erasure. Extensive experiments demonstrate that SAFER consistently and effectively erases unwanted concepts from diffusion models while preserving generation quality. © 2025, CC BY.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Reverse-Nearest-Neighbor-Based Clustering by Fast Search and Find of Density Peaks

引用

Chinese Journal of Electronics 2023年第6期32卷 1341-1354页

作者： ZHANG Chunhao XIE Bin ZHANG Yiran College of Computer and Cyber Security Hebei Normal University Hebei Provincial Engineering Research Center for Supply Chain Big Data Analytics and Data Security Hebei Normal University Hebei Provincial Key Laboratory of Network and Information Security Hebei Normal University

Clustering by fast search and find of density peaks(CFSFDP) has the advantages of a novel idea, easy implementation, and efficient clustering. It has been widely recognized in various fields since it was proposed in Science in 2014. The CFSFDP algorithm also has certain limitations, such as non-unified sample density metrics defined by cutoff distance, the domino effect for the assignment of remaining samples triggered by unstable assignment strategy, and the phenomenon of picking wrong density peaks as cluster centers. We propose reverse-nearest-neighbor-based clustering by fast search and find of density peaks(RNN-CFSFDP) to avoid these shortcomings. We redesign and unify the sample density metric by introducing reverse nearest neighbor. The newly defined local density metric and the K-nearest neighbors of each sample are combined to make the assignment process more robust and alleviate the domino effect. A cluster fusion algorithm is proposed, which further alleviates the domino effect and effectively avoids the phenomenon of picking wrong density peaks as cluster centers. Experimental results on publicly available synthetic data sets and real-world data sets show that in most cases, the proposed algorithm is superior to or at least equivalent to the comparative methods in clustering performance. The proposed algorithm works better on manifold data sets and uneven density data sets.

关键词： Measurement Manifolds Shape Clustering algorithms Synthetic data

来源：评论

学校读者我要写书评

暂无评论

The bug that made me president a browser- and web-security case study on helios voting

The bug that made me president a browser- and web-security c...

引用

3rd International Conference on E-Voting and Identity, VoteID 2011

作者： Heiderich, Mario Frosch, Tilman Niemietz, Marcus Schwenk, Jörg Department of Network and Data Security Ruhr-University Bochum Germany

ISBN: (纸本)9783642327469

This paper briefly describes security challenges for critical web applications such as the Helios Voting system. After analyzing the Helios demonstration website we discovered several small flaws that can have a large security critical impact. An attacker is able to extract sensitive information, manipulate voting results, and modify the displayed information of Helios without any deep technical knowledge or laboratory-like prerequisites. Displaying and processing trusted information in an untrustworthy user agent can lead to the issue that most protection mechanisms are useless. In our approach of attacking Helios voting systems we do not rely on an already infected or trojanized machine of the user, instead we use simple and commonly known web browser features to leverage information disclosure and state modification attacks. We propose that online voting applications should at least follow the latest vulnerability mitigation guidelines. In addition, there should be thorough and frequent coverage with automated as well as manual penetrations tests in privacy sensitive applications. E-Voting software driven by web browsers is likely to become an attractive target for attackers. Successful exploitation can have impact ranging from large scale personal information leakage, financial damage, calamitously intended information and state modification as well as severe real life impact in many regards. © 2012 Springer-Verlag.

关键词： Web browsers

来源：评论

学校读者我要写书评

暂无评论

DRL-Based Time-Varying Workload Scheduling With Priority and Resource Awareness

引用

IEEE Transactions on network and Service Management 2025年

作者： Liu, Qifeng Fan, Qilin Zhang, Xu Li, Xiuhua Wang, Kai Xiong, Qingyu Chongqing University School of Big Data and Software Engineering Chongqing400044 China Chongqing University Key Laboratory of Dependable Service Computing in Cyber Physical Society of Ministry of Education Chongqing400044 China Nanjing University School of Electronic Science and Engineering Nanjing210023 China Haihe Laboratory of Information Technology Application Innovation Tianjin300072 China Harbin Institute of Technology School of Computer Science and Technology Weihai264209 China Shandong Key Laboratory of Industrial Network Security Weihai264209 China

With the proliferation of cloud services and the continuous growth in enterprises' demand for dynamic multi-dimensional resources, the implementation of effective strategy for time-varying workload scheduling has become increasingly significant. In this paper, we propose a deep reinforcement learning (DRL)-based method for time-varying workload scheduling, aiming to allocate resources efficiently across servers in the cluster. Specifically, we integrate a classifier and queue scorer to construct a priority queue that exploits temporal resource utilization patterns across different workload classes. Then, we design parallel graph attention layers to capture the dimensional features and temporal dynamics of cloud server cluster. Moreover, we propose a DRL algorithm to generate scheduling strategies that can adapt to dynamic environments. Validation on real-world traces from Google cluster demonstrates that our method outperforms existing approaches in key metrics of cloud server cluster management. © 2004-2012 IEEE.

关键词： Cloud platforms

来源：评论

学校读者我要写书评

暂无评论

CrossLinkNet: An Explainable and Trustworthy AI Framework for Whole-Slide Images Segmentation

引用

Computers, Materials & Continua 2024年第6期79卷 4703-4724页

作者： Peng Xiao Qi Zhong Jingxue Chen Dongyuan Wu Zhen Qin Erqiang Zhou Network and Data Security Key Laboratory of Sichuan Province University of Electronic Science and Technology of ChinaChengdu610054China Faculty of Data Science City University of MacaoMacao999078China

In the intelligent medical diagnosis area,Artificial Intelligence(AI)’s trustworthiness,reliability,and interpretability are critical,especially in cancer *** neural networks,while excellent at processing natural images,often lack interpretability and adaptability when processing high-resolution digital pathological *** limitation is particularly evident in pathological diagnosis,which is the gold standard of cancer diagnosis and relies on a pathologist’s careful examination and analysis of digital pathological slides to identify the features and progression of the ***,the integration of interpretable AI into smart medical diagnosis is not only an inevitable technological trend but also a key to improving diagnostic accuracy and *** this paper,we introduce an innovative Multi-Scale Multi-Branch Feature Encoder(MSBE)and present the design of the CrossLinkNet *** MSBE enhances the network’s capability for feature extraction by allowing the adjustment of hyperparameters to configure the number of branches and *** CrossLinkNet Framework,serving as a versatile image segmentation network architecture,employs cross-layer encoder-decoder connections for multi-level feature fusion,thereby enhancing feature integration and segmentation *** quantitative and qualitative experiments on two datasets demonstrate that CrossLinkNet,equipped with the MSBE encoder,not only achieves accurate segmentation results but is also adaptable to various tumor segmentation tasks and scenarios by replacing different feature ***,CrossLinkNet emphasizes the interpretability of the AI model,a crucial aspect for medical professionals,providing an in-depth understanding of the model’s decisions and thereby enhancing trust and reliability in AI-assisted diagnostics.

关键词： Explainable AI security trustworthy CrossLinkNet whole slide images

来源：评论

学校读者我要写书评

暂无评论

Revisiting SGX-based Encrypted Deduplication via PoW-Before-Encryption and Eliminating Redundant Computations

引用

IEEE Transactions on Dependable and Secure Computing 2024年第3期22卷 2037-2053页

作者： Ha, Guanxiong Ge, Xiaowei Jia, Chunfu Chen, Yuchen Su, Zhen Nankai University Tianjin Key Laboratory of Network and Data Security Technology DISSEC College of Cyber Science Tianjin300350 China

Encrypted deduplication is attractive for outsourced storage as it provides both data confidentiality and storage savings. Conventional encrypted deduplication schemes protect data confidentiality based on expensive cryptographic primitives, leading to performance degradation. Recently, several SGX-based schemes have been proposed to accelerate encrypted deduplication. However, these schemes have limitations in both security and performance aspects. This paper presents a SGX-based basic scheme to address these limitations, which first performs proof of ownership (PoW), followed by key generation and data encryption, realizing a new paradigm known as PoW-before-encryption (PbE) to solve the security issue in existing schemes. Additionally, the basic scheme implements deduplication-before-encryption (DbE) to reduce redundant computations, thus improving performance. Despite these improvements, the duplicate detection and key generation in the basic scheme still involve redundant computations. Consequently, we propose an epoch-based enhanced scheme that utilizes data locality and computation deduplication, which caches fresh computations in an epoch and reuses them to enhance performance. We provide a security analysis and evaluate the performance of our schemes using both synthetic and real-world workloads. The results demonstrate that our schemes offer stronger security guarantees while outperforming state-of-the-art schemes in terms of performance. © 2004-2012 IEEE.

关键词： Cryptography

来源：评论

学校读者我要写书评

暂无评论

Poster: A footprint of third-party tracking on mobile web 13

Poster: A footprint of third-party tracking on mobile web

引用

2013 ACM SIGSAC Conference on Computer and Communications security, CCS 2013

作者： Javed, Ashar Department of Network and Data Security Ruhr University Bochum Bochum Germany

ISBN: (纸本)9781450324779

In this paper, we investigate the footprints of third-party tracking on the mobile web. The survey of 100 popular mobile versions of web applications indicates that third-party tracking is also prevalent on mobile web. The results show that 62 sites are tracking users' activities on mobile web and Google Analytics is the most widespread tracker on mobile web. We believe that this study will help raise awareness about the tracking situation on the mobile web side. © 2013 Author.

关键词： Surveys

来源：评论

学校读者我要写书评

暂无评论

AdaptFRCNet: Semi-supervised adaptation of pre-trained model with frequency and region consistency for medical image segmentation

引用

Medical Image Analysis 2025年 103卷 103626页

作者： He, Along Wu, Yanlin Wang, Zhihong Li, Tao Fu, Huazhu College of Computer Science Tianjin Key Laboratory of Network and Data Security Technology Nankai University Tianjin300350 China 138632 Singapore Haihe Lab of ITAI Tianjin300459 China

Recently, large pre-trained models (LPM) have achieved great success, which provides rich feature representation for downstream tasks. Pre-training and then fine-tuning is an effective way to utilize LPM. However, the application of LPM in the medical domain is hindered by the presence of a large number of parameters and a limited amount of labeled data. In clinical practice, there exists a substantial amount of unlabeled data that remains underutilized. Semi-supervised learning emerges as a promising approach to harnessing these unlabeled data effectively. In this paper, we propose semi-supervised adaptation of pre-trained model with frequency and region consistency (AdaptFRCNet) for medical image segmentation. Specifically, the pre-trained model is frozen and the proposed lightweight attention-based adapters (Att_Adapter) are inserted into the frozen backbone for parameter-efficient fine-tuning (PEFT). We propose two consistency regularization strategies for semi-supervised learning: frequency domain consistency (FDC) and multi-granularity region similarity consistency (MRSC). FDC aids in learning features within the frequency domain, and MRSC aims to achieve multiple region-level feature consistencies, capturing local context information effectively. By leveraging the proposed Att_Adapter along with FDC and MRSC, we can effectively and efficiently harness the powerful feature representation capability of the LPM. We conduct extensive experiments on three medical image segmentation datasets, demonstrating significant performance improvements over other state-of-the-art methods. The code is available at https://***/NKUhealong/AdaptFRCNet. © 2025 Elsevier B.V.

关键词： Semi-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Correctness verification in database outsourcing: A trust-based fake tuples approach

Correctness verification in database outsourcing: A trust-ba...

引用

8th International Conference on Information Systems security, ICISS 2012

作者： Ghasemi, Simin Noferesti, Morteza Hadavi, Mohammad Ali Nogoorani, Sadegh Dorri Jalili, Rasool Data and Network Security Lab Sharif University of Technology Tehran Iran

ISBN: (纸本)9783642351297

An important security challenge in database outsourcing scenarios is the correctness verification of query results. The proposed approaches in the literature, impose high overhead on both the service provider and specially the clients. In this paper, we propose the Trust-Based Fake Tuples approach to audit the correctness of query results. In this approach, some fake tuples are included among the real ones in order to verify the correctness of the results. The experience learnt from past results is used in this paper to evaluate the trust toward the service provider. This trust value is used to tune the number of fake tuples and subsequently the imposed overhead. As the trust value toward the service provider increases, the number of fake tuples and imposed overhead decreases. The experimental results confirm the effectiveness of our approach on the reduction of overhead while keeping the probability of incorrect results in its acceptable level. © Springer-Verlag 2012.

关键词： Outsourcing

来源：评论

学校读者我要写书评

暂无评论

Modelling simultaneous mutual authentication for authenticated key exchange

Modelling simultaneous mutual authentication for authenticat...

引用

6th International Symposium on Foundations and Practice of security, FPS 2013

作者： Yang, Zheng Horst Görtz Institute for IT Security Department for Network- and Data Security Ruhr-University Bochum Germany

ISBN: (纸本)9783319053011

Most recent security models for authenticated key exchange (AKE) do not explicitly model the entity authentication, which enables a party to identify its communication peer in specific session. However, it is quite necessary in many real-world applications and is a general way to enhance the security of AKE protocols. Despite much work on AKE, we notice that there is no good definition of entity authentication security involving simultaneous protocol execution that would improve the bandwidth efficiency in practice. Based on eCK model, we define a security model called eCK-A that deals with simultaneous mutual authentication. Besides the eCK-A model particularly formulates the security properties regarding resilience to the leakage of various combinations of long-term key and ephemeral session state, and provision of perfect forward secrecy in a single model. We present a generic protocol compiler to achieve the eCK-A security based on any eCK secure protocols. © 2014 Springer International Publishing Switzerland.

关键词： Authentication

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：