检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Wang, Ziqiu Liu, Jun Zhang, Shengkai Yang, Yang Wuhan University of Technology Wuhan China Key Laboratory of Intelligent Sensing System and Security Ministry of Education School of Artificial Intelligence Hubei University Wuhan China

With the development of Natural Language Processing (NLP), Large Language Models (LLMs) are becoming increasingly popular. LLMs are integrating more into everyday life, raising public concerns about their security vulnerabilities. Consequently, the security of large language models is becoming critically important. Currently, the techniques for attacking and defending against LLMs are continuously evolving. One significant method type of attack is the jailbreak attack, which designed to evade model safety mechanisms and induce the generation of inappropriate content. Existing jailbreak attacks primarily rely on crafting inducement prompts for direct jailbreaks, which are less effective against large models with robust filtering and high comprehension abilities. Given the increasing demand for real-time capabilities in large language models, real-time updates and iterations of new knowledge have become essential. Retrieval-Augmented Generation (RAG), an advanced technique to compensate for the model’s lack of new knowledge, is gradually becoming mainstream. As RAG enables the model to utilize external knowledge bases, it provides a new avenue for jailbreak attacks. In this paper, we conduct the first work to propose the concept of indirect jailbreak and achieve Retrieval-Augmented Generation via LangChain. Building on this, we further design a novel method of indirect jailbreak attack, termed Poisoned-LangChain (PLC), which leverages a poisoned external knowledge base to interact with large language models, thereby causing the large models to generate malicious non-compliant *** tested this method on six different large language models across three major categories of jailbreak issues. The experiments demonstrate that PLC successfully implemented indirect jailbreak attacks under three different scenarios, achieving success rates of 88.56%, 79.04%, and 82.69% respectively. Experimental results and other resources: https://***/CAM-FSS/jailbreak-langchai

关键词： Knowledge based systems

来源：评论

学校读者我要写书评

暂无评论

Stacking U-Nets in U-shape: Redesigning the Information Flow in Model-based Networks for MRI Reconstruction

Stacking U-Nets in U-shape: Redesigning the Information Flow...

引用

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

作者： Xiaoyu Qiao Weisheng Li Bin Xiao Yuping Huang Lijian Yang Chongqing Key Laboratory of Image Cognition Chongqing University of Posts and Telecommunications Chongqing China Key Laboratory of Cyberspace Big Data Intelligent Security Chongqing University of Posts and Telecommunications Ministry of Education Chongqing China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Model-based networks have shown convincing performance in MRI reconstruction. However, the unrolled cascades within the networks are constrained to solely obtain information from the preceding counterpart, resulting in potential error accumulation. Moreover, the linear structure fails to address the challenge of recovering fine-grained details. To tackle these problems, we propose to redesign the information flow in model-based networks. Our method features a large U-shaped network, where the nodes are built with unrolled cascades and U-Net-based regularizers. We design an input-level integration module to help the cascades acquire information from adjacent and skip-connected counterparts, building robust mappings to the target. We further design a coarse-to-fine feature-level integration module, aiming at guiding the network to progressively recover fine details. Intermediate reconstructions produced by subnetworks of different scales are integrated, enabling the extraction of complementary information to enhance the final performance. Compared with cutting-edge methods on different datasets, our method exhibits superior performances.

关键词： Magnetic resonance imaging Stacking Buildings Signal processing Feature extraction Acoustics Sensors data mining Speech processing

来源：评论

学校读者我要写书评

暂无评论

Several new infinite families of NMDS codes with arbitrary dimensions supporting t-designs

arXiv

引用

arXiv 2025年

作者： Zhang, Yaozong Zheng, Dabin Wang, Xiaoqiang Lu, Wei Hubei Key Laboratory of Applied Mathematics Faculty of Mathematics and Statistics Hubei University Wuhan430062 China Key Laboratory of Intelligent Sensing System and Security Ministry of Education Hubei University Wuhan430062 China

Near maximum distance separable (NMDS) codes, where both the code and its dual are almost maximum distance separable, play pivotal roles in combinatorial design theory and cryptographic applications. Despite progress in fixed dimensions (e.g., dimension 4 codes by Ding and Tang [5]), constructing NMDS codes with arbitrary dimensions supporting t-designs (t ≥ 2) has remained open. In this paper, we construct two infinite families of NMDS codes over Fq for any prime power q with flexible dimensions and determine their weight distributions. Further, two additional families with arbitrary dimensions over F2m supporting 2-designs and 3-designs, and their weight distributions are obtained. Our results fully generalize prior fixed-dimension works [8], [16], [17], [32], and affirmatively settle the Heng-Wang conjecture [16] on the existence of NMDS codes with flexible parameters supporting 2-designs. Copyright © 2025, The Authors. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Learning Preconditioners in Gates-controlled Deep Unfolding Networks based on Quasi-Newton Methods For Accelerated MRI Reconstruction

Learning Preconditioners in Gates-controlled Deep Unfolding ...

引用

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

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Deep unfolding networks (DUNs) have made significant progress in MRI reconstruction, successfully tackling the problem of prolonged imaging time. However, the ill-conditioned nature of MRI reconstruction often causes slow convergence in iterative optimization, potentially compromising the performance of DUNs. In this study we propose a preconditioned and gates-controlled DUN (PGDUN) to address these challenges. Our approach starts with optimizing the step size of proximal gradient descent (PGD) through a preconditioner. To improve flexibility and adaptability, we relax the constrains on quasi-newton-based optimization procedure. We design ConvLSTM-based modules, where the gate units automatically preserve necessary long- and short-term information, facilitating the learning of optimized variables and their combinations. Furthermore, we design gate units to modulate the features fed to regularizers across different iterations, boosting their robustness against potential accumulated errors. Evaluations using PSNR and SSIM metrics reveal that our approach outperforms existing state-of-the-art methods, achieving superior reconstruction results across various sequences.

关键词： Measurement Magnetic resonance imaging Logic gates Signal processing Robustness Sensors Iterative methods Speech processing Image reconstruction Optimization

来源：评论

学校读者我要写书评

暂无评论

A New Dual-Population Evolutionary Algorithm Leveraging Objective-Constraint Relationships for Constrained Multi-Objective Optimization

SSRN

引用

SSRN 2025年

作者： Ye, Jialu Tan, Chaogui Xia, Yizhang Hou, Zhanglu Liu, Yuan Zou, Juan Hunan Engineering Research Center of intelligent System Optimization and Security Key Laboratory of Intelligent Computing and Information Processing Ministry of Education Hunan National Applied Mathematics Center China Xiangtan University China

Population co-evolution strategies are widely used to handle constrained multi-objective optimization problems (CMOPs). However, existing coevolutionary algorithms oversimplify population collaboration and are rigid in evolving the auxiliary population, ignoring the importance of maintaining population diversity, especially for CMOPs with irregular feasible regions. To tackle this issue, we propose a new dual-population evolutionary algorithm, denoted as AACMO, which attempts to mine the relational features between the unconstrained Pareto front (UPF) and the constrained Pareto front (CPF) to better adapt to CMOPs with various constraints. Specifically, the method is categorized into two phases: a learning phase and an evolving phase. First, the main population (MP) aims to quickly learn the relationships between the UPF and the CPF by leveraging the advantages of different operators, namely, genetic algorithm operators and differential evolution operators, during the learning phase. Secondly, two auxiliary strategies are designed to collaborate with the MP during the evolving phase. These strategies are capable of employing a more effective co-evolution method to assist the MP in maintaining diversity, based on the degree of alignment between the learned UPF and CPF at the end of the learning phase. The effectiveness of AACMO is confirmed through comparative analysis with eight state-of-the-art algorithms across four CMOP benchmark suites. © 2025, The Authors. All rights reserved.

关键词： Optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Generating Is Believing: Membership Inference Attacks against Retrieval-Augmented Generation

Generating Is Believing: Membership Inference Attacks agains...

引用

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

作者： Yuying Li Gaoyang Liu Chen Wang Yang Yang Key Laboratory of Intelligent Sensing System and Security (Ministry of Education) School of Artificial Intelligence Hubei University Wuhan China Hubei Key Laboratory of Smart Internet Technology Huazhong University of Science and Technology Wuhan China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Retrieval-Augmented Generation (RAG) is a state-of-the-art technique that mitigates issues such as hallucinations and knowledge staleness in Large Language Models (LLMs) by retrieving relevant knowledge from an external database to assist in content generation. Existing research has demonstrated potential privacy risks associated with the LLMs of RAG. However, the privacy risks posed by the integration of an external database, which often contains sensitive data such as medical records or personal identities, have remained largely unexplored. In this paper, we aim to bridge this gap by focusing on membership privacy of RAG’s external database, with the aim of determining whether a given sample is part of the RAG’s database. Our basic idea is that if a sample is in the external database, it will exhibit a high degree of semantic similarity to the text generated by the RAG system. We present S 2 MIA, a Membership Inference Attack that utilizes the Semantic Similarity between a given sample and the content generated by the RAG system. With our proposed S 2 MIA, we demonstrate the potential to breach the membership privacy of the RAG database. Extensive experimental results demonstrate that S 2 MIA outperforms five existing MIAs, even when the system is protected by three representative defenses.

关键词： data privacy Privacy databases Large language models Semantics Retrieval augmented generation Focusing Signal processing Solids Speech processing

来源：评论

学校读者我要写书评

暂无评论

Rack-aware MSR Codes with Optimal Access for Multiple Sequentially Ordered Node Failures

引用

IEEE Transactions on Communications 2025年

作者： Wang, Jiaojiao Zheng, Dabin Guan, Xun Tsinghua University Tsinghua Shenzhen International Graduate School Guangdong Shenzhen518055 China Hubei University Hubei Key Laboratory of Applied Mathematics Faculty of Mathematics and Statistics Key Laboratory of Intelligent Sensing System and Security Ministry of Education Wuhan430062 China

The minimum storage rack-aware regenerating (MSRR) code is a variation of regenerating codes that achieves the optimal repair bandwidth for a single node failure within the rack-aware model. We study the access complexity of repairing MSRR codes, that allows collective information processing among nodes within the same rack. A previous study has reported construction of MSRR codes that require accessing the minimum number of symbols to repair a single node. We extend this work by constructing a family of MSRR codes that minimizes the number of symbols accessed to repair sequentially ordered failed nodes in a single rack, and further show that for certain code parameters, another version of MSRR codes can be constructed with reduced sub-packetization while still preserving the optimal access property. © 1972-2012 IEEE.

关键词： Turbo codes

来源：评论

学校读者我要写书评

暂无评论

GENERATING IS BELIEVING: MEMBERSHIP INFERENCE ATTACKS AGAINST RETRIEVAL-AUGMENTED GENERATION

arXiv

引用

arXiv 2024年

作者： Li, Yuying Liu, Gaoyang Wang, Chen Yang, Yang Key Laboratory of Intelligent Sensing System and Security Ministry of Education School of Artificial Intelligence Hubei University Wuhan China Huazhong University of Science and Technology Wuhan China

Retrieval-Augmented Generation (RAG) is a state-of-the-art technique that mitigates issues such as hallucinations and knowledge staleness in Large Language Models (LLMs) by retrieving relevant knowledge from an external database to assist in content generation. Existing research has demonstrated potential privacy risks associated with the LLMs of RAG. However, the privacy risks posed by the integration of an external database, which often contains sensitive data such as medical records or personal identities, have remained largely unexplored. In this paper, we aim to bridge this gap by focusing on membership privacy of RAG’s external database, with the aim of determining whether a given sample is part of the RAG’s database. Our basic idea is that if a sample is in the external database, it will exhibit a high degree of semantic similarity to the text generated by the RAG system. We present S2MIA, a Membership Inference Attack that utilizes the Semantic Similarity between a given sample and the content generated by the RAG system. With our proposed S2MIA, we demonstrate the potential to breach the membership privacy of the RAG database. Extensive experiment results demonstrate that S2MIA can achieve a strong inference performance compared with five existing MIAs, and is able to escape from the protection of three representative defenses. © 2024, CC BY.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

A Contribution Assessment Method Based on Model Performance Gains in Federated Learning

A Contribution Assessment Method Based on Model Performance ...

引用

International Symposium on Parallel and Distributed Processing with Applications, ISPA

作者： Zixuan Zhao Jun Liu Yuyin Li Gaoyang Liu Yamin Li Yang Yang Key Laboratory of Intelligent Sensing System and Security (Ministry of Education) School of Artificial Intelligence Hubei University Wuhan China Huazhong University of Science and Technology Wuhan China

ISBN: (数字)9798331509712

ISBN: (纸本)9798331509729

Federated Learning is a distributed model training system that uses the computing resources and private data of participants to collaboratively train Machine Learning models. In order to incentivize data holders to actively participate in Federated Learning, a crucial issue is how to fairly assess each participant’s contribution. One class of the mainstream methods estimates client contributions by calculating the correlation between local and global model parameters, substantially reducing the reliance on large computing resources and test datasets. However, these methods overlook an implicit key factor, leading to inaccuracy in assessing the contribution of the participants. We propose FedCA, which provides a fairness assessment of the client’s contribution based on the performance gain of the global model in each communication round. The experimental results demonstrate that FedCA accurately identifies the contributions of clients with varying data qualities and effectively approximates the optimal fairness method of Shapley values. The source code for this paper is available at https://***/paper-liu/***.

关键词： Training Distributed processing Correlation Federated learning Computational modeling Source coding data integrity Distributed databases Performance gain data models

来源：评论

学校读者我要写书评

暂无评论

Score-based Generative Priors Guided Model-driven Network for MRI Reconstruction

arXiv

引用

arXiv 2024年

作者： Qiao, Xiaoyu Li, Weisheng Xiao, Bin Huang, Yuping Yang, Lijian Chongqing Key Laboratory of Image Cognition Chongqing University of Posts and Telecommunications Chongqing China Key Laboratory of Cyberspace Big Data Intelligent Security Chongqing University of Posts and Telecommunications Ministry of Education Chongqing China

Score matching with Langevin dynamics (SMLD) method has been successfully applied to accelerated MRI. However, the hyperparameters in the sampling process require subtle tuning, otherwise the results can be severely corrupted by hallucination artifacts, especially with out-of-distribution test data. To address the limitations, we proposed a novel workflow where naive SMLD samples serve as additional priors to guide model-driven network training. First, we adopted a pretrained score network to generate samples as preliminary guidance images (PGI), obviating the need for network retraining, parameter tuning and in-distribution test data. Although PGIs are corrupted by hallucination artifacts, we believe they can provide extra information through effective denoising steps to facilitate reconstruction. Therefore, we designed a denoising module (DM) in the second step to coarsely eliminate artifacts and noises from PGIs. The features are extracted from a score-based information extractor (SIE) and a cross-domain information extractor (CIE), which directly map to the noise patterns. Third, we designed a model-driven network guided by denoised PGIs (DGIs) to further recover fine details. DGIs are densely connected with intermediate reconstructions in each cascade to enrich the information and are periodically updated to provide more accurate guidance. Our experiments on different datasets reveal that despite the low average quality of PGIs, the proposed workflow can effectively extract valuable information to guide the network training, even with severely reduced training data and sampling steps. Our method outperforms other cutting-edge techniques by effectively mitigating hallucination artifacts, yielding robust and high-quality reconstruction results. Copyright © 2024, The Authors. All rights reserved.

关键词： Compressed sensing

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：