检索结果-内蒙古大学图书馆

1st Conference on Blockchain and Web3.0 Technology Innovation and Application, BWTAC 2024

作者： Tang, Hao Wang, Cheng Tongji University Shanghai201804 China The Key Laboratory of Embedded System and Service Computing Ministry of Education Shanghai201804 China Shanghai Artificial Intelligence Laboratory Shanghai200030 China Shanghai Institute of Intelligent Science and Technology Tongji University Shanghai200092 China

ISBN: (纸本)9789819794119

Social behavior metadata has altered business models and human lifestyles. However, the inclusion of personal information in social behavior metadata poses risks of identity exposure for users. Under the requirements of modern data management laws, service providers now employ anonymization techniques when collecting user behavior metadata, which aims to break the link between personal identifiers and behavioral attributes for privacy protection. This paper conducts experiments on an open-source dataset of anonymized social behavior metadata and finds that even with privacy protection techniques such as data generalization, there remains a high risk of reidentification. Specifically, four social behavior metadata records are sufficient to reidentify users from the anonymized social behavior metadata, achieving a unicity of 90.10%. This indicates that the current widespread collection of social behavior metadata brings a high risk of privacy breaches. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Anonymity

来源：评论

学校读者我要写书评

暂无评论

Dual Intention Escape: Penetrating and Toxic Jailbreak Attack against Large Language Models 25

Dual Intention Escape: Penetrating and Toxic Jailbreak Attac...

引用

Proceedings of the ACM on Web Conference 2025

作者： Yanni Xue Jiakai Wang Zixin Yin Yuqing Ma Haotong Qin Renshuai Tao Xianglong Liu State Key Laboratory of Complex & Critical Software Environment Beihang University Beijing China Zhongguancun Laboratory Beijing Beijing China Beihang University Beijing China Institute of Artificial Intelligence Beihang University Beijing China ETH Zurich Zurich Switzerland Institute of Information Science Beijing Jiaotong University Beijing China State Key Laboratory of Complex & Critical Software Environment Beihang University Beijing China Zhongguancun Laboratory Beijing China and Institute of Data Space Hefei Comprehensive National Science Center Hefei China

ISBN: (纸本)9798400712746

Recently, the jailbreak attack, which generates adversarial prompts to bypass safety measures and mislead large language models (LLMs) to output harmful answers, has attracted extensive interest due to its potential to reveal the vulnerabilities of LLMs. However, ignoring the exploitation of the characteristics in intention understanding, existing studies could only generate prompts with weak attacking ability, failing to evade defenses (e.g., sensitive word detect) and causing malice(e.g., harmful outputs). Motivated by the mechanism in the psychology of human misjudgment, we propose a dual intention escape (DIE) jailbreak attack framework to generate more stealthy and toxic prompts to deceive LLMs to output harmful content. For stealthiness, inspired by the anchoring effect, we designed the Intention-anchored Malicious Concealment(IMC) module that hides the harmful intention behind a generated anchor intention by the recursive decomposition block and contrary intention nesting block. Since the anchor intention will be received first, the LLMs might pay less attention to the harmful intention and enter response status. For toxicity, we propose the Intention-reinforced Malicious Inducement (IMI) module based on the availability bias mechanism in a progressive malicious prompting approach. Due to the ongoing emergence of statements correlated to harmful intentions, the output content of LLMs will be closer to these more accessible intentions, i.e., more toxic. We conducted extensive experiments under black-box settings, supporting that DIE could achieve 100% ASR-R and 92.9% ASR-G against GPT3.5-turbo.

关键词： dual intention escape

来源：评论

学校读者我要写书评

暂无评论

Continual Test-Time Adaptation for Single Image Defocus Deblurring via Causal Siamese Networks

arXiv

引用

arXiv 2025年

作者： Cui, Shuang Li, Yi Li, Jiangmeng Tang, Xiongxin Su, Bing Xu, Fanjiang Xiong, Hui National Key Laboratory of Space Integrated Information System Institute of Software Chinese Academy of Sciences University of Chinese Academy of Sciences China Beijing Key Laboratory of Big Data Management and Analysis Methods Gaoling School of Artificial Intelligence Renmin University of China China Department of Computer Science & Engineering the Hong Kong University of Science and Technology Hong Kong

Single image defocus deblurring (SIDD) aims to restore an all-in-focus image from a defocused one. Distribution shifts in defocused images generally lead to performance degradation of existing methods during out-of-distribution inferences. In this work, we gauge the intrinsic reason behind the performance degradation, which is identified as the heterogeneity of lens-specific point spread functions. Empirical evidence supports this finding, motivating us to employ a continual test-time adaptation (CTTA) paradigm for SIDD. However, traditional CTTA methods, which primarily rely on entropy minimization, cannot sufficiently explore task-dependent information for pixel-level regression tasks like SIDD. To address this issue, we propose a novel Siamese networks-based continual test-time adaptation framework, which adapts source models to continuously changing target domains only requiring unlabeled target data in an online manner. To further mitigate semantically erroneous textures introduced by source SIDD models under severe degradation, we revisit the learning paradigm through a structural causal model and propose Causal Siamese networks (CauSiam). Our method leverages large-scale pre-trained vision-language models to derive discriminative universal semantic priors and integrates these priors into Siamese networks, ensuring causal identifiability between blurry inputs and restored images. Extensive experiments demonstrate that CauSiam effectively improves the generalization performance of existing SIDD methods in continuously changing domains. Copyright © 2025, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

CESFusion: Cross-Frequency Enhanced Spatial-Spectral Fusion Network for Hyperspectral and Multispectral Image Fusion

引用

IEEE Transactions on Geoscience and Remote Sensing 2025年 63卷

作者： Zhang, Haozheng Yang, Yanhong Li, Chaoyang Lu, Yanjie Zhang, Guodao Chen, Shengyong School of Computer Science and Engineering Tianjin University of Technology Tianjin Key Laboratory of Intelligence Computing and Novel Software Technology Tianjin300384 China Engineering Research Center of Learning- Based Intelligent System Ministry of Education Key Laboratory of Computer Vision and Systems School of Computer Science and Engineering Tianjin University of Technology Tianjin300384 China Institute of Intelligent Media Computing Hangzhou Dianzi University Hangzhou310018 China

The fusion of hyperspectral and multispectral images involves integrating high spectral resolution hyperspectral image (HSI) and high spatial resolution multispectral image (MSI) to generate a HSI with high spatial and spectral resolution (HR-HSI). Existing HSI-MSI fusion methods primarily focus on information fusion within the spatial domain;however, few solutions have explored the employment of frequency analysis to enhance spatial resolution, limiting their capability for global perception. In this paper, we propose an efficient and novel paradigm for HSI-MSI fusion through the cross-frequency enhanced spatial-spectral fusion network, named CESFusion, exploring the complementary fusion of information between the spatial and frequency domains. Specifically, we first present the cross-frequency domain fusion module (CFFM) to perform global analysis through the Fourier transform and effectively integrate and enhance the frequency domain information from both HSI and MSI. Subsequently, we propose the spectral modeling module (SpeMM) based on state space model (SMM) to capture long-range spectral dependencies with linear complexity, and integrate it with the spatial residual block-based module (SRM) for joint spatial-spectral feature extraction. Finally, to enable sufficient interaction between the spatial and frequency domains, we adopt the cross-domain interaction module (CDIM), capturing and integrating complementary information from both domains. Moreover, a frequency-based loss function is purposely designed to further improve the restoration of global information. Extensive experiments conducted on both synthetic and real datasets demonstrate the superiority of our CESFusion, as evidenced by both quantitative and qualitative evaluation results. © 2025 IEEE.

关键词： Image fusion

来源：评论

学校读者我要写书评

暂无评论

High Efficiency Wiener Filter-based Point Cloud Quality Enhancement for MPEG G-PCC

引用

IEEE Transactions on Circuits and systems for Video Technology 2025年

作者： Wei, Yuxuan Wang, Zehan Guo, Tian Liu, Hao Shen, Liquan Yuan, Hui Shandong University School of Control Science and Engineering Jinan250061 China Ministry of Education Key Laboratory of Machine Intelligence and System Control Jinan250061 China Yantai University School of Computer and Control Engineering Yantai264005 China Shanghai University Shanghai Institute for Advanced Communication and Data Science Shanghai200072 China

Point clouds, which directly record the geometry and attributes of scenes or objects by a large number of points, are widely used in various applications such as virtual reality and immersive communication. However, due to the huge data volume and unstructured geometry, efficient compression of point clouds is very crucial. The Moving Picture Expert Group is establishing a geometry-based point cloud compression (G-PCC) standard for both static and dynamic point clouds in recent years. Although lossy compression of G-PCC can achieve a very high compression ratio, the reconstruction quality is relatively low, especially at low bitrates. To mitigate this problem, we propose a high efficiency Wiener filter that can be integrated into the encoder and decoder pipeline of G-PCC to improve the reconstruction quality as well as the rate-distortion performance for dynamic point clouds. Specifically, we first propose a basic Wiener filter, and then improve it by introducing coefficients inheritance and variance-based point classification for the Luma component. Besides, to reduce the complexity of the nearest neighbor search during the application of the Wiener filter, we also propose a Morton code-based fast nearest neighbor search algorithm for efficient calculation of filter coefficients. Experimental results demonstrate that the proposed method can achieve average Bjøntegaard delta rates of -6.1%, -7.3%, and -8.0% for Luma, Chroma Cb, and Chroma Cr components, respectively, under the condition of lossless-geometry-lossy-attributes configuration compared to the latest G-PCC encoding platform (i.e., geometry-based solid content test model version 7.0 release candidate 2) by consuming affordable computational complexity. © 2025 IEEE. All rights reserved.

关键词： Wiener filtering

来源：评论

学校读者我要写书评

暂无评论

Perceptual Uncertainty-Aware Motion Planning for Autonomous Driving Based on Adaptive Heuristic Reinforcement Learning

引用

IEEE Transactions on Intelligent Transportation systems 2025年

作者： Hu, Chuan Du, Baiyu Wang, Zhidong Zhang, Qichao Zhang, Ronghui Gao, Hongbo Shanghai Jiao Tong University School of Mechanical Engineering Shanghai200240 China Chinese Academy of Sciences State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Beijing100190 China University of Chinese Academy of Sciences School of Artificial Intelligence Beijing100049 China Sun Yat-sen University Guangdong Provincial Key Laboratory of Intelligent Transport System School of Intelligent Systems Engineering Guangzhou510275 China University of Science and Technology of China School of Information Science and Technology Department of Automation Hefei230026 China

Autonomous driving has become an inevitable trend in automotive development. Reinforcement Learning (RL) is extensively used in autonomous vehicle motion planning, demonstrating good generalization but facing challenges of long training times and lack of consideration of uncertainty. To address these challenges, an Adaptive Heuristic Reinforcement Learning (AHRL) approach is proposed. First, this study improves upon the Dueling Double Deep Q Network (D3QN) algorithm by proposing the Adaptive Heuristic Dueling Double Deep Q Network (Adapt-HD3QN) algorithm. Specifically, heuristic functions from search-based planning algorithms are incorporated into the RL reward terms to guide heuristic learning and enhance learning efficiency. Additionally, considering the uncertainties in real-world driving environments, such as the movement of other traffic participants and building occlusions, a Mixed Artificial Potential Field (Mix-APF) is implemented to address static and dynamic obstacle avoidance. Furthermore, potential collisions between the autonomous vehicle and other vehicles in occluded areas are modeled as a zero-sum game, with a Dynamic Bayesian Network (DBN) used for prior modeling of potential vehicles, aiding in constructing the potential vehicle’s forward hidden set. Finally, a signal-free intersection scenario, a typical crash-prone road type, is constructed on CARLA, incorporating static obstacles and other traffic participants. Experimental results demonstrate that the proposed Adapt-HD3QN algorithm exhibits superior safety, training efficiency, and traffic efficiency in scenarios with road environment perception uncertainties caused by occlusion. © 2000-2011 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A Triangular Stable Node Network based on Self-supervised Learning for personalized prediction

A Triangular Stable Node Network based on Self-supervised Le...

引用

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

作者： Liu, Qing Gao, Qian Fan, Jun Zhang, Zhiqiang Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center National Supercomputer Center in Jinan Qilu University of Technology Shandong Academy of Sciences Jinan250014 China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Qilu University of Technology Shandong Academy of Sciences Jinan250353 China Shandong Provincial Key Laboratory of Industrial Network and Information System Security Shandong Fundamental Research Center for Computer Science Jinan250014 China China Telecom Digital Intelligence Techonology Co Ltd. No.1999 Shunhua road Shandong Jinan250101 China

ISBN: (纸本)9798350368741

In recent years, research has illuminated the potency of implicit data processing in enhancing user preferences. Nevertheless, barriers remain in breaking through the constraints of implicit information. This study aims to bridge this gap by firstly constructing a triangular stable node network model, tailored to manage implicit information with precision. Recognizing the challenge of pinpointing novel structures within large-scale graphs, we propose SLTSNN-a triangular stable node network based on self-supervised learning for personalized prediction. SLTSNN innovates by maximizing mutual information between graph-level and patch-level representations, while augmenting graph representations through extended enhanced representations. Additionally, it incorporates triangular data associations and introduces a triangular allocation attention network, which emphasizes strongly correlated preference features among similar users. Furthermore, SLTSNN employs contrastive learning to maximize mutual information between graph vectors and hidden representations, distinguishing between high-order global and local representations. The model's effectiveness in enhancing user preferences and capturing new graph structures is evidenced by its performance on hit rate and normalized discounted cumulative gain metrics across three datasets. © 2025 IEEE.

关键词： graph contrastive learning graph neural networks recommender systems self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Moving-Window Integrated Physics-Data-Based Modeling of Four-Wheel Vehicle Lateral Dynamics

引用

Automotive Innovation 2025年 1-11页

作者： Wei, Wenpeng Wang, Jinxiang Pi, Dawei Zhuang, Weichao He, Tianyi Yin, Guodong School of Mechanical Engineering Southeast University Nanjing China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Southeast University Nanjing China Safety and Intelligence of Vehicle Equipment Research Center Nanjing University of Science and Technology Nanjing China Department of Mechanical and Aerospace Engineering Utah State University Logan USA

This paper presents an Integrated physics-data-based (IPDB) modeling of lateral vehicle dynamics with moving-window data snapshots. The IPDB model encodes the fundamental physical principle of four-wheel vehicle motions and simultaneously carries out the adaptiveness of the data-driven approach. Specifically, the traditional physics-based lateral dynamics considering four-wheel interaction are first derived into an affine linear-parameter-varying model, in which the vehicle-related parameters and motion variables are separated. Then, by using the Kronecker product, the IPDB model, directly formulated by the data snapshots in the moving-window fashion, is obtained for system representation. As a result, the IPDB technique rendered model is physically interpretable. The impacts of moving window length on modeling performance are numerically studied. The IPDB model accuracy is validated with data from CarSim simulations and experiments with passenger vehicles under various scenarios. It is further demonstrated that the proposed IPDB model is more data-efficient than other data-driven methods since it only uses the data snapshot in the moving window to update the model recursively. This characteristic enables the IPDB method with online modeling capability to adapt to varying driving scenarios.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Unleashing Correlation and Continuity for Hyperspectral Reconstruction from RGB Images

arXiv

引用

arXiv 2025年

作者： Feng, Fuxiang Cong, Runmin Wei, Shoushui Zhang, Yipeng Li, Jun Kwong, Sam Zhang, Wei The School of Control Science and Engineering Shandong University Jinan250061 China The Key Laboratory of Machine Intelligence and System Control Ministry of Education Jinan250061 China The University of California Los Angeles United States The Hubei Key Laboratory of Intelligent Geo-Information Processing The School of Computer Science China University of Geo-sciences Wuhan430078 China The Lingnan University Hong Kong

Reconstructing Hyperspectral Images (HSI) from RGB images can yield high spatial resolution HSI at a lower cost, demonstrating significant application potential. This paper reveals that local correlation and global continuity of the spectral characteristics are crucial for HSI reconstruction tasks. Therefore, we fully explore these inter-spectral relationships and propose a Correlation and Continuity Network (CCNet) for HSI reconstruction from RGB images. For the correlation of local spectrum, we introduce the Group-wise Spectral Correlation Modeling (GrSCM) module, which efficiently establishes spectral band similarity within a localized range. For the continuity of global spectrum, we design the Neighborhood-wise Spectral Continuity Modeling (NeSCM) module, which employs memory units to recursively model the progressive variation characteristics at the global level. In order to explore the inherent complementarity of these two modules, we design the Patch-wise Adaptive Fusion (PAF) module to efficiently integrate global continuity features into the spectral features in a patch-wise adaptive manner. These innovations enhance the quality of reconstructed HSI. We perform comprehensive comparison and ablation experiments on the mainstream datasets NTIRE2022 and NTIRE2020 for the spectral reconstruction task. Compared to the current advanced spectral reconstruction algorithms, our designed algorithm achieves State-Of-The-Art (SOTA) performance. Copyright © 2025, The Authors. All rights reserved.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Replay Without Saving: Prototype Derivation and Distribution Rebalance for Class-Incremental Semantic Segmentation

引用

IEEE Transactions on Pattern Analysis and Machine intelligence 2025年第6期47卷 4699-4716页

作者： Jinpeng Chen Runmin Cong Yuxuan Luo Horace Ho Shing Ip Sam Kwong Department of Computer Science City University of Hong Kong Kowloon Hong Kong School of Control Science and Engineering Shandong University Jinan China Key Laboratory of Machine Intelligence and System Control Ministry of Education Jinan China Department of Computer Science and Centre for Innovative Applications of Internet and Multimedia Technologies City University of Hong Kong Kowloon Hong Kong School of Data Science Lingnan University Tuen Mun Hong Kong

The research of class-incremental semantic segmentation (CISS) seeks to enhance semantic segmentation methods by enabling the progressive learning of new classes while preserving knowledge of previously learned ones. A significant yet often neglected challenge in this domain is class imbalance. In CISS, each task focuses on different foreground classes, with the training set for each task exclusively comprising images that contain these currently focused classes. This results in an overrepresentation of these classes within the single-task training set, leading to a classification bias towards them. To address this issue, we propose a novel CISS method named STAR, whose core principle is to reintegrate the missing proportions of previous classes into current single-task training samples by replaying their prototypes. Moreover, we develop a prototype deviation technique that enables the deduction of past-class prototypes, integrating the recognition patterns of the classifiers and the extraction patterns of the feature extractor. With this technique, replay can be accomplished without using any storage to save prototypes. Complementing our method, we devise two loss functions to enforce cross-task feature constraints: the Old-Class Features Maintaining (OCFM) loss and the Similarity-Aware Discriminative (SAD) loss. The OCFM loss is designed to stabilize the feature space of old classes, thus preserving previously acquired knowledge without compromising the ability to learn new classes. The SAD loss aims to enhance feature distinctions between similar old and new class pairs, minimizing potential confusion. Our experiments on two public datasets, Pascal VOC 2012 and ADE20 K, demonstrate that our STAR achieves state-of-the-art performance.

关键词： Prototypes Training Feature extraction Semantic segmentation Continuing education Stars Semantics Vectors Transformers Image reconstruction

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：