检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Xu, Huiyu Zhang, Wenhui Wang, Zhibo Xiao, Feng Zheng, Rui Feng, Yunhe Ba, Zhongjie Ren, Kui The State Key Laboratory of Blockchain and Data Security Zhejiang University China School of Cyber Science and Technology Zhejiang University China Palo Alto Networks United States Department of Computer Science and Engineering University of North Texas United States

Recently, advanced Large Language Models (LLMs) such as GPT-4 have been integrated into many real-world applications like Code Copilot. These applications have significantly expanded the attack surface of LLMs, exposing them to a variety of threats. Among them, jailbreak attacks that induce toxic responses through carefully constructed jailbreak prompts have raised critical safety concerns. To effectively identify these threats, a growing number of red team approaches simulate potential adversarial scenarios by crafting jailbreak prompts to test the target LLM. However, existing red teaming methods do not consider the unique vulnerabilities of LLM in different scenarios (e.g., code-related tasks), making it difficult to adjust the jailbreak prompts to find context-specific vulnerabilities, thereby lacking efficiency. Meanwhile, these methods are limited to refining handcrafted jailbreak templates using a few mutation operations (such as synonym replacement), lacking the automation and scalability to continuously adapt to different scenarios. To enable context-aware and efficient red teaming, we abstract and model existing attacks into a coherent concept called "jailbreak strategy" and propose a multi-agent LLM system named RedAgent that leverages these strategies to generate context-aware jailbreak prompts. By self-reflecting on contextual feedback and red teaming trials in an additional memory buffer, RedAgent continuously learns how to leverage these strategies to achieve more effective jailbreaks in specific contexts. Extensive experiments demonstrate that our system can jailbreak most black-box LLMs within just five queries, improving the efficiency of existing red teaming methods by two times. Additionally, RedAgent can jailbreak customized LLM applications more efficiently. By generating context-aware jailbreak prompts towards trending applications on GPTs, we discover 60 severe vulnerabilities of these real-world applications with only two queries per vulnera

关键词： Autonomous agents

来源：评论

学校读者我要写书评

暂无评论

Online Discriminative Semantic-Preserving Hashing for Large-Scale Cross-Modal Retrieval 18th

Online Discriminative Semantic-Preserving Hashing for Large-...

引用

18th Pacific Rim International Conference on Artificial Intelligence, PRICAI 2021

作者： Yi, Jinhan He, Yi Liu, Xin Department of Computer Science and Technology Huaqiao University Xiamen361021 China Key Laboratory of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education Nanjing University of Science and Technology Nanjing210094 China Xiamen Key Laboratory of Computer Vision and Pattern Recognition Xiamen China Fujian Key Laboratory of Big Data Intelligence and Security Xiamen China

ISBN: (纸本)9783030891879

Cross-modal hashing has drawn increasing attentions for efficient retrieval across different modalities, and existing methods primarily learn the hash functions in a batch based mode, i.e., offline methods. Nevertheless, the multimedia data often comes in a streaming fashion, which makes the batch based learning methods uncompetitive for large-scale streaming data due to the large memory consumption and calculation. To address this problem, we present an Online Discriminative Semantic-Preserving Hashing (ODSPH) method for large-scale cross-modal retrieval, featuring on fast training speed, low memory consumption and high retrieval accuracy. Within the proposed ODSPH framework, we utilize the newly coming data points to learn the hash codes and update hash functions in a stream manner. When new data comes, the corresponding hash codes are obtained by regressing the class label of the training examples. For hash function, we update it with the accumulated information from each round. Besides, we design a novel momentum updating method to adaptively update the hash function and reduce quantization loss, which can produce discriminative hash codes for high retrieval precision. Extensive experiments on three benchmark datasets show that the proposed ODSPH method improves the retrieval performance over the state-of-the-arts. © 2021, Springer Nature Switzerland AG.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

A Rational Delegating Computation Protocol Based on Smart Contract 14th

A Rational Delegating Computation Protocol Based on Smart Co...

引用

14th EAI International Conference on Mobile Multimedia Communications, Mobimedia 2021

作者： Ma, Juan Chen, Yuling Liu, Guoxu Zhu, Hongliang Peng, Changgen State Key Laboratory of Public Big Data College of Computer Science and Technology Guizhou University Guiyang China Blockchain Laboratory of Agricultural Vegetables Weifang University of Science and Technology Shouguang China Engineering Lab for Cloud Security and Information Security Center of BUPT Beijing China

ISBN: (纸本)9783030898137

The delegating computation has become an irreversible trend, together comes the pressing need for fairness and efficiency issues. To solve this problem, we leverage game theory to propose a smart contract-based solution. Firstly, according to the behavioral preferences of the participants, we design an incentive contract to describe the motivation of the participants. Secondly, to satisfy the fairness of the rational delegating computation, we propose a delegating computation protocol based on the smart contract. Specifically, rational participants are to gain the maximum utility and reach the Nash equilibrium in the protocol. Besides, we design a reputation mechanism with a reputation certificate, which measures the reputation from multiple dimensions. And the reputation is used to assure the client’s trust in the computing party to improve the efficiency of the protocol. Finally, the analysis results show that the proposed protocol solves the complex traditional verification problem. Meanwhile, we prove the fairness and correctness of the protocol. © 2021, ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.

关键词： Smart contract

来源：评论

学校读者我要写书评

暂无评论

Code Search Is All You Need? Improving Code Suggestions with Code Search

Code Search Is All You Need? Improving Code Suggestions with...

引用

International Conference on Software Engineering (ICSE)

作者： Junkai Chen Xing Hu Zhenhao Li Cuiyun Gao Xin Xia David Lo School of Software Technology Zhejiang University Ningbo China The State Key Laboratory of Blockchain and Data Security Zhejiang University Hangzhou China Concordia University Montreal Canada Harbin Institute of Technology Shenzhen China Zhejiang University Hangzhou China Singapore Management University Singapore

ISBN: (数字)9798400702174

ISBN: (纸本)9798350382143

Modern integrated development environments (IDEs) provide various automated code suggestion techniques (e.g., code completion and code generation) to help developers improve their efficiency. Such techniques may retrieve similar code snippets from the code base or leverage deep learning models to provide code suggestions. However, how to effectively enhance the code suggestions using code retrieval has not been systematically investigated. In this paper, we study and explore a retrieval-augmented framework for code suggestions. Specifically, our framework leverages different retrieval approaches and search strategies to search similar code snippets. Then the retrieved code is used to further enhance the performance of language models on code suggestions. We conduct experiments by integrating different language models into our framework and compare the results with their original models. We find that our framework noticeably improves the performance of both code completion and code generation by up to 53.8% and 130.8% in terms of BLEU-4, respectively. Our study highlights that integrating the retrieval process into code suggestions can improve the performance of code suggestions by a large margin.

关键词： Deep learning Codes Search problems Software engineering

来源：评论

学校读者我要写书评

暂无评论

On Optimizing Inter-and Intra-Chiplet Interconnection Topologies for Robust Multi-Chiplet Systems

引用

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 2025年

作者： Wang, Xiaohang Xu, Miao Singh, Amit Kumar Jiang, Yingtao Yang, Mei Zhejiang University State Key Laboratory of Blockchain and Data Security China Hangzhou China South China University of Technology School of Software Engineering China University of Essex School of Computer Science and Electronic Engineering United Kingdom University of Nevada Department of Electrical and Computer Engineering Las Vegas United States

Inter-and intra-chiplet interconnection networks play a vital role in the operation of many core systems made of multiple chiplets. However, these networks are susceptible to faults caused by manufacturing defects and attacks resulting from the malicious insertion of hardware Trojans and backdoors. Unlike conventional fault-tolerant or countermeasure methods, this paper focuses on optimizing network robustness to withstand both faults and attacks, while considering the constraints of chiplet area and power budget. To achieve this, this paper first defines network robustness as a quantifiable measure based on various network parameters, after which an optimization problem is formulated to optimize the robustness of the network topology. To efficiently solve this problem, a reinforcement learning algorithm is proposed. Experimental results demonstrate that the proposed method is capable of generating inter-and intra-chiplet interconnection networks that are significantly more robust than existing topology generation methods. Specifically, the proposed method improves robustness over ButterDonut and Kite, respectively by an average of 10.88% and 14.06% under random faults and by 9.37% and 7.81% under targeted attacks. These experimental results confirm that the proposed method is capable of generating robust inter-and intra-chiplet interconnection networks that can withstand both faults and attacks. By optimizing the network topology’s robustness, it provides a valuable contribution to the design and security of chiplet-based core systems. © 2025 IEEE. All rights reserved

关键词： Budget control

来源：评论

学校读者我要写书评

暂无评论

Optimizing Potato Disease Detection in Pakistan with Machine Learning: A Comparative Analysis

Optimizing Potato Disease Detection in Pakistan with Machine...

引用

Emerging Trends in Electrical, Control, and Telecommunication Engineering (ETECTE), International Conference on

作者： Irfan Ahmed Manahal Khalil Safoora Ghulam Jabar Mahmood Khadija Batool Nibtal Khalid Pakiza Arshad Dept of Computer Science University of Sialkot Sialkot Pakistan State Key Laboratory of Blockchain and Data Security School of Cyber Science and Technology College of Computer Science & Technology Zhejiang University Hangzhou China Dept of Computer Science Int. Institute of Science Art & Technology Gujranwala Pakistan

ISBN: (数字)9798331504861

ISBN: (纸本)9798331504878

The agricultural sector of Pakistan depends heavily on the production of potatoes, however diseases like Bacterial Wilt, Late Blight, and Early Blight are posing a growing danger to this industry since they can negatively affect crop productivity and food security. Detecting these illnesses using traditional methods is time-consuming, labor-intensive, and sometimes it requires specialized knowledge which is not easy for farmers. The implementation of a machine learning (ML) framework for illness identification using the methods of K-Nearest Neighbours (KNN), Random Forest (RF), Decision Tree (DT), Naive Bayes (NB), Support Vector Machine (SVM), and Logistic Regression (LR) overcomes these challenges. This study utilizes data mining and preprocessing strategies to improve the accuracy of models and adaptability on a dataset of potato plants’ photos gathered across diverse areas in Pakistan. The models are assessed using performance criteria such computational efficiency, accuracy, precision, recall and F1-score. Random Forest model having 97% accuracy outperforms the other ML models in identifying plant disease. The suggested approach provides a practical, affordable, and effective way to identify diseases in a timely manner, allowing farmers to make informed decisions to lower crop losses. The aim of the study is to highlight how ML has the potential to transform Pakistani agriculture, enhancing food security and promoting long-term economic prosperity.

关键词： Support vector machines Productivity Adaptation models Accuracy Biological system modeling Computational modeling Crops Food security Random forests Classification tree analysis

来源：评论

学校读者我要写书评

暂无评论

Dbfl: Dynamic Byzantine-Robust Privacy Preserving Federated Learning in Heterogeneous data Scenario

SSRN

引用

SSRN 2024年

作者： Chen, Xiaoli Tian, Youliang Wang, Shuai Yang, Kedi Zhao, Wei Xiong, Jinbo State Key Laboratory of Public Big Data College of Computer Science and Technology Guizhou University Guizhou Guiyang550025 China Guizhou Provincial Key Laboratory of Cryptography and Blockchain Technology Guizhou University Guizhou Guiyang550025 China College of Mathematics and Statistics Guizhou University Guizhou Guiyang550025 China Fujian Provincial Key Laboratory of Network Security and Cryptology College of Computer and Cyber Security Fujian Normal University Fujian Fuzhou350117 China

Privacy Preserving Federated Learning (PPFL) protects the clients' local data privacy by securely uploading encrypted gradients to the server. However, in real-world scenarios, heterogeneous client data distribution makes identifying poisoning gradients challenging. And during local iterations, models moving in different directions constantly shift the boundary between benign and malicious gradients. To address these challenges, we design a Dynamic Byzantine-robust defense strategy (DBFL) based on two-trapdoor homomorphic encryption, achieving detection of encrypted model poisoning attacks. Specifically, we first propose a secure Manhattan Distance method, which accurately measures the difference between each element in two encrypted gradients, thus enabling precise detection of poisoning attacks in heterogeneous data scenarios without exposing any privacy. Furthermore, we design a Byzantine-tolerant aggregation mechanism based on dynamic threshold, where the dynamic threshold changes according to the continuously changing boundary between poisoning and benign gradients in heterogeneous data scenarios. Even when malicious clients controlled by Byzantine attackers constitute 70\% of the DBFL, our method accurately excludes poisoned gradients. security analysis demonstrates that DBFL achieves IND-CPA security. Extensive evaluation on two benchmark datasets shows that DBFL outperforms existing defense strategies. In particular, DBFL achieves a 15\%-40\% accuracy improvement in the non-IID setting compared to existing solutions. © 2024, The Authors. All rights reserved.

关键词： Network security

来源：评论

学校读者我要写书评

暂无评论

Dynamic Graph Transformer with Correlated Spatial-Temporal Positional Encoding

arXiv

引用

arXiv 2024年

作者： Wang, Zhe Zhou, Sheng Chen, Jiawei Zhang, Zhen Hu, Binbin Feng, Yan Chen, Chun Wang, Can Zhejiang University College of Computer Science Hangzhou China National University of Singapore Singapore Zhejiang University School of Software Technology Ningbo China Ant Group Hangzhou China Zhejiang University The State Key Laboratory of Blockchain and Data Security Hangzhou China

Learning effective representations for Continuous-Time Dynamic Graphs (CTDGs) has garnered significant research interest, largely due to its powerful capabilities in modeling complex interactions between nodes. A fundamental and crucial requirement for representation learning in CTDGs is the appropriate estimation and preservation of proximity. However, due to the sparse and evolving characteristics of CTDGs, the spatial-temporal properties inherent in high-order proximity remain largely unexplored. Despite its importance, this property presents significant challenges due to the computationally intensive nature of personalized interaction intensity estimation and the dynamic attributes of CTDGs. To this end, we propose a novel Correlated Spatial-Temporal Positional encoding that incorporates a parameter-free personalized interaction intensity estimation under the weak assumption of the Poisson Point Process. Building on this, we introduce the Dynamic Graph Transformer with Correlated Spatial-Temporal Positional Encoding (CorDGT), which efficiently retains the evolving spatial-temporal high-order proximity for effective node representation learning in CTDGs. Extensive experiments on seven small and two large-scale datasets demonstrate the superior performance and scalability of the proposed CorDGT. The code is available at: https://***/wangz3066/CorDGT. © 2024, CC BY.

关键词： Encoding (symbols)

来源：评论

学校读者我要写书评

暂无评论

Autoregressive Queries for Adaptive Tracking with Spatio-Temporal Transformers

Autoregressive Queries for Adaptive Tracking with Spatio-Tem...

引用

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Jinxia Xie Bineng Zhong Zhiyi Mo Shengping Zhang Liangtao Shi Shuxiang Song Rongrong Ji Key Laboratory of Education Blockchain and Intelligent Technology Ministry of Education Guangxi Normal University Guilin China Guangxi Key Lab of Multi-Source Information Mining & Security Guangxi Normal University Guilin China Guangxi Key Laboratory of Machine Vision and Intelligent Control Wuzhou University School of Computer Science and Technology Harbin Institute of Technology Media Analytics and Computing Lab School of Informatics Xiamen University

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

The rich spatio-temporal information is crucial to capture the complicated target appearance variations in visual tracking. However, most top-performing tracking algorithms rely on many hand-crafted components for spatio-temporal information aggregation. Consequently, the spatio-temporal information is far away from being fully explored. To alleviate this issue, we propose an adaptive tracker with spatio-temporal transformers (named AQA-Track), which adopts simple autoregressive queries to effectively learn spatio-temporal information without many hand-designed components. Firstly, we introduce a set of learnable and autoregressive queries to capture the instantaneous target appearance changes in a sliding window fashion. Then, we design a novel attention mechanism for the interaction of existing queries to generate a new query in current frame. Finally, based on the initial target template and learnt autoregressive queries, a spatio-temporal information fusion module (STM) is designed for spatiotemporal formation aggregation to locate a target object. Benefiting from the STM, we can effectively combine the static appearance and instantaneous changes to guide robust tracking. Extensive experiments show that our method significantly improves the tracker's performance on six popular tracking benchmarks: LaSOT, LaSOT ext , TrackingNet, GOT-10k, TNL2K, and UAV123. Code and models will be https://***/orgs/GXNU-ZhongLab.

关键词： Adaptation models Visualization Computer vision Target tracking Computational modeling Transformers Market research

来源：评论

学校读者我要写书评

暂无评论

Autoregressive Queries for Adaptive Tracking with Spatio-Temporal Transformers

arXiv

引用

arXiv 2024年

作者： Xie, Jinxia Zhong, Bineng Mo, Zhiyi Zhang, Shengping Shi, Liangtao Song, Shuxiang Ji, Rongrong Key Laboratory of Education Blockchain and Intelligent Technology Ministry of Education China Guangxi Key Lab of Multi-Source Information Mining & Security China Guangxi Normal University Guilin541004 China Guangxi Key Laboratory of Machine Vision and Intelligent Control Wuzhou University China School of Computer Science and Technology Harbin Institute of Technology China Media Analytics and Computing Lab School of Informatics Xiamen University China

The rich spatio-temporal information is crucial to capture the complicated target appearance variations in visual tracking. However, most top-performing tracking algorithms rely on many hand-crafted components for spatio-temporal information aggregation. Consequently, the spatio-temporal information is far away from being fully explored. To alleviate this issue, we propose an adaptive tracker with spatio-temporal transformers (named AQATrack), which adopts simple autoregressive queries to effectively learn spatio-temporal information without many hand-designed components. Firstly, we introduce a set of learnable and autoregressive queries to capture the instantaneous target appearance changes in a sliding window fashion. Then, we design a novel attention mechanism for the interaction of existing queries to generate a new query in current frame. Finally, based on the initial target template and learnt autoregressive queries, a spatio-temporal information fusion module (STM) is designed for spatio-temporal formation aggregation to locate a target object. Benefiting from the STM, we can effectively combine the static appearance and instantaneous changes to guide robust tracking. Extensive experiments show that our method significantly improves the tracker’s performance on six popular tracking benchmarks: LaSOT, LaSOText, TrackingNet, GOT-10k, TNL2K, and *** and models will be here. Copyright © 2024, The Authors. All rights reserved.

关键词： Benchmarking

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：