检索结果-内蒙古大学图书馆

11th International Conference on Machine Intelligence Theory and Applications, MiTA 2024

作者： Quadri, Hakeem Gu, Bruce Wang, Hua Islam, Sardar Victoria University Department of Information Technology Melbourne Australia Qilu University of Technology Key Laboratory of Computing Power Network and Information Security Jinan China Victoria University Institute for sustainable Industries and Liveable Cities Melbourne Australia

ISBN: (纸本)9798350385670

Convolutional Neural networks (CNN), particularly low latency models such as MobileNets have excelled in many applications, including object detection in images, speech recognition and natural language processing, but they are vulnerable to subtle perturbations that are virtually imperceptible to the human eye yet can deceive the network into misclassifying images. To enhance the robustness of such CNNs against adversaries, conventional adversarial training methods treat all data points as equally important and susceptible to attack. A weighted adversarial learning algorithm is developed in a Stackelberg game framework. This approach prioritizes data points that are more susceptible to attacks during network training. To further optimize the algorithm, we employ a Reinforcement Learning (R L) agent to fine-tune the hyperparameters of the model, thereby increasing its robustness. Our findings indicate an increased robustness of 66.18 % of the Weighted Adversary Reinforced Stackelberg Learning (WARS) against the traditional adversarial training of 64.72 % in a one epoch training, using the CIFAR-10 dataset. We conclude that the WARS represents a valuable adversarial training method for bolstering the robustness of low latency CNN models. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

PosFuzz:augmenting greybox fuzzing with effective position distribution

引用

Cybersecurity 2023年第4期6卷 123-143页

作者： Yanyan Zou Wei Zou JiaCheng Zhao Nanyu Zhong Yu Zhang Ji Shi Wei Huo Institute of Information Engineering Chinese Academy of SciencesBeijingChina School of Cyber Security University of Chinese Academy of SciencesBeijingChina State Key Lab of Processors Institute of Computing TechnologyChinese Academy of SciencesBeijingChina Key Laboratory of Network Assessment Technology Chinese Academy of SciencesBeijingChina Beijing Key Laboratory of Network Security and Protection Technology BeijingChina Zhongguancun Laboratory BeijingChina

Mutation-based greybox fuzzing has been one of the most prevalent techniques for security vulnerability discovery and a great deal of research work has been proposed to improve both its efficiency and ***-based greybox fuzzing generates input cases by mutating the input seed,i.e.,applying a sequence of mutation operators to randomly selected mutation positions of the ***,existing fruitful research work focuses on scheduling mutation operators,leaving the schedule of mutation positions as an overlooked aspect of fuzzing *** paper proposes a novel greybox fuzzing method,PosFuzz,that statistically schedules mutation positions based on their historical *** makes use of a concept of effective position distribution to represent the semantics of the input and to guide the *** first utilizes Good-Turing frequency estimation to calculate an effective position distribution for each mutation *** then leverages two sampling methods in different mutating stages to select the positions from the *** have implemented PosFuzz on top of AFL,AFLFast and MOPT,called Pos-AFL,-AFLFast and-MOPT respectively,and evaluated them on the UNIFUZZ benchmark(20 widely used open source programs)and LAVA-M *** result shows that,under the same testing time budget,the Pos-AFL,-AFLFast and-MOPT outperform their counterparts in code coverage and vulnerability discovery *** with AFL,AFLFast,and MOPT,PosFuzz gets 21%more edge coverage and finds 133%more paths on *** also triggers 275%more unique bugs on average.

关键词： Greybox fuzzing Mutation position Mutation operator Code coverage Vulnerability discovery

来源：评论

学校读者我要写书评

暂无评论

JTEA: Implementing Semi-Supervised Entity Alignment Using Joint Teaching Strategies

JTEA: Implementing Semi-Supervised Entity Alignment Using Jo...

引用

2024 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2024

作者： Lu, Kai Zhao, Jing Ding, Lichao Hao, Zenghao Key Laboratory of Computing Power Network and Information Security Ministry of Education Jinan China Faculty of Computer Science and Technology Jinan China Shandong Fundamental Research Center for Computer Science Shandong Provincial Key Laboratory of Computer Networks Jinan China

ISBN: (纸本)9781665410205

Entity alignment(EA) aims to identify equivalent entities in different knowledge graphs or data sources and establish correspondences between them. In recent years, embedding-based methods have made significant progress in the field of entity alignment, yet the shortage of adequate training data (i.e., seed entity pairs) continues to be a primary challenge. Traditional semi-supervised methods typically address this issue by generating potential mappings (i.e., pseudo-mappings) for unlabeled entities. However, these methods often suffer from the influence of erroneous potential mappings or overlook the uncertainty of potential mappings. Therefore, we propose a novel semi-supervised entity alignment method called JTEA. This method combines seed entity pairs with potential mappings through end-to-end joint training, providing strong guidance for model training. In constructing potential mappings, we propose a bidirectional matching strategy (BMS), which generates more detailed potential mappings by jointly adjusting decisions in different directions and uses a joint matching confidence score to characterize its uncertainty. Furthermore, we also introduce a method of matching diversity alignment selection, aiming to screen out more reliable entity alignments to further improve the performance and stability of the model. We conducted a large number of experiments on the benchmark dataset to fully validate the effectiveness of the JTEA method. © 2024 IEEE.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

Dynamic detection technology of encryption algorithm based on Pin 9

Dynamic detection technology of encryption algorithm based o...

引用

9th IEEE International Conference on Data Science in Cyberspace, DSC 2024

作者： Qu, Junpeng Du, Gaolei Sun, Yunge Fu, Yong Shandong Computer Science Center National Supercomputer Center in Jinan Qilu University of Technology Shandong Academy of Sciences Key Laboratory of Computing Power Network and Information Security Ministry of Education Jinan250014 China Quan Cheng Laboratory Jinan250103 China Shandong Fundamental Research Center for Computer Science Shandong Provincial Key Laboratory of Industrial Network and Information System Security Jinan China

ISBN: (纸本)9798350391367

Aiming at the problem that malicious software uses encryption algorithm to hide its own malicious code and avoid detection, a dynamic detection technology based on Pin encryption algorithm is proposed. Use Pin to get the instruction flow of executable file running, then identify the function call in the instruction flow to get the function call framework, find the loop body in the calling function, and then carry out the encryption algorithm instruction detection and encryption algorithm constant detection on the loop body. If the detection passes, it can be considered that a component of the encryption algorithm is found. Then the complete encryption algorithm is found according to the generated function call framework. Experiments show that this method can effectively identify the encryption algorithms hidden in executable files. © 2024 IEEE.

关键词： Encryption algorithms

来源：评论

学校读者我要写书评

暂无评论

An Encryption Algorithm Cycle Identification Method Based on Bit Execution 9

An Encryption Algorithm Cycle Identification Method Based on...

引用

9th IEEE International Conference on Data Science in Cyberspace, DSC 2024

作者： Sun, Yunge Du, Gaolei Qu, Junpeng Fu, Yong Qilu University of Technology Shandong Academy of Sciences Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center National Supercomputer Center in Jinan Jinan250014 China Quan Cheng Laboratory Jinan250103 China Shandong Fundamental Research Center for Computer Science Shandong Provincial Key Laboratory of Industrial Network and Information System Security Jinan China

ISBN: (纸本)9798350391367

Ransomware uses symmetric encryption algorithms to lock user files to extort money. This paper proposes a cryptographic algorithm loop identification method based on bit-precise execution, aiming to solve the problem of identifying unknown loops. The structure of the loop body of encryption algorithms is fixed and usually includes nonlinear operations and linear operations. This article converts the loop body in the unknown software into a set of Boolean formulas, and converts the malware into a register data flow through the symbolic executor, thereby converting it into a set of Boolean formulas for execution. Execute known encryption algorithms with bit accuracy and establish a Boolean formula input-output relationship library for cryptographic components. The two sets of Boolean formulas are passed through a designed input matrix to achieve the Boolean formula mapping and input variable mapping of the confirmation algorithm, which can effectively judge the encryption algorithms in some malware. © 2024 IEEE.

关键词： Encryption algorithms

来源：评论

学校读者我要写书评

暂无评论

Enhancing Stance Detection on Social Media via Core Views Discovery 27

Enhancing Stance Detection on Social Media via Core Views Di...

引用

27th European Conference on Artificial Intelligence, ECAI 2024

作者： Yan, Yu Shen, Yinghan Liu, Teli Jiang, Xuhui Yin, Dechun School of Information and Network Security People Public Security University of China China Institute of Computing Technology Chinese Academy of Sciences China Key Laboratory of AI Safety Institute of Computing Technology Chinese Academy of Sciences China

ISBN: (纸本)9781643685489

Stance detection aims to identify the expressed attitude towards a target from the text, which is significant for learning public cognition from social media. The short and implicit nature of social media users' expressions potentially results in the stance understanding bias of the model. To address this problem, introducing external background information is helpful to mitigate these biases and enhance explainability. The core view, reflecting the motivations and reasons behind an individual's stance toward the target, can be summarized and extracted from collective tweets, which can serve as a reference for stance detection. In this study, we propose the Stance Detection via Core View Discovery (SD-CVM), where the core views are used for background information modeling. Specifically, we construct a joint classifier combining the semantic understanding of tweets and their relevant core views from the public. We utilize the Large Language Model (LLM) to extract core views with stances from tweets and use these core views as background references for tweets. To further optimize the tweet understanding, we develop the contrastive and rebalancing mechanism by incorporating stance supervision signals for training. Experiments on two representative datasets demonstrate the excellent performance of our method. © 2024 The Authors.

关键词： Tweets

来源：评论

学校读者我要写书评

暂无评论

An automated construction method for symmetric encryption round function models based on dynamic analysis 9

An automated construction method for symmetric encryption ro...

引用

9th IEEE International Conference on Data Science in Cyberspace, DSC 2024

作者： Du, Gaolei Sun, Yunge Qu, Junpeng Fu, Yong Qilu University of Technology Shandong Academy of Sciences Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center National Supercomputer Center in Jinan Jinan250014 China Quan Cheng Laboratory Jinan250103 China Shandong Fundamental Research Center for Computer Science Shandong Provincial Key Laboratory of Industrial Network and Information System Security Jinan China

ISBN: (纸本)9798350391367

With the deepening of informatization, the incidence of cybercrimes utilizing malicious software for attacks is on the rise. This paper addresses the detection issue of symmetric encryption in malicious software from the perspective of cryptography, proposing an automated method for constructing a symmetric encryption round function model based on dynamic analysis. This method tracks data flow at the granularity of registers to characterize Directed Acyclic Graph (DAG) construction of round function models. Initially, dynamic execution traces of binary files are captured using Intel Pin, identifying loops within them. Data flow tracking at the register level within loops is performed, creating DAG, and analyzing whether the loops exhibit characteristics of symmetric cipher round functions, thus constructing the round function model. This approach enables the detection of symmetric encryption within binary files and the construction of round function models, offering a novel solution for fine-grained symmetric cipher recognition and laying a foundation for detecting malicious software containing symmetric encryption. © 2024 IEEE.

关键词： Data flow graphs

来源：评论

学校读者我要写书评

暂无评论

Self-Evaluation of Large Language Model based on Glass-box Features

Self-Evaluation of Large Language Model based on Glass-box F...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Huang, Hui Qu, Yingqi Liu, Jing Yang, Muyun Xu, Bing Zhao, Tiejun Lu, Wenpeng Faculty of Computing Harbin Institute of Technology Harbin China Baidu Inc. Beijing China Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology Jinan China

ISBN: (纸本)9798891761681

The proliferation of open-source Large Language Models (LLMs) underscores the pressing need for evaluation methods. Existing works primarily rely on external evaluators, focusing on training and prompting strategies. However, a crucial aspect - model-aware glass-box features - is overlooked. In this study, we explore the utility of glass-box features under the scenario of self-evaluation, namely applying an LLM to evaluate its own output. We investigate various glass-box feature groups and discovered that the softmax distribution serves as a reliable quality indicator for self-evaluation. Experimental results on public benchmarks validate the feasibility of self-evaluation of LLMs using glass-box features. © 2024 Association for Computational Linguistics.

关键词： Open systems

来源：评论

学校读者我要写书评

暂无评论

Generating Personalized Imputations for Patient Health Status Prediction in Electronic Health Records

Generating Personalized Imputations for Patient Health Statu...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Liu, Yang Zhang, Weiyu Si, Jiasheng Li, Zhao Peng, Xueping Lu, Wenpeng Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Jinan China Evay Info Jinan China Shandong Fundamental Research Center for Computer Science Shandong Provincial Key Laboratory of Computing Power Internet and Service Computing Jinan China University of Technology Sydney Australian Artificial Intelligence Institute Sydney Australia

ISBN: (纸本)9798350386226

Electronic health records (EHRs) play a crucial role in the development of personalized treatment plans for patients. However, EHRs are often highly incomplete, posing significant challenges for predictive modeling. While existing deep learning models employ various imputation techniques to reconstruct missing values, they often fail to represent missing data in a personalized manner and do not learn from the missingness patterns in EHRs data. This limitation reduces their effectiveness in practical personalized healthcare applications. To address this issue, we propose SPIME, a self-supervised model that generates personalized imputations in EHRs data for patient health status prediction. We introduce a personalized missing mask (PMM) based on the frequency of feature measurements. Additionally, we incorporate a masked imputation task (MIT) loss that minimizes the loss of artificially introduced missing values, thereby enhancing the model's capability to handle missing data. SPIME adopts self-supervised pretraining to learn representations from personalized missing patterns and reconstructs missing data in the latent space. To further enhance representation learning, two independent attention mechanisms are applied separately across the feature and temporal dimensions. Experimental results on two real-world EHRs datasets show that SPIME outperforms existing state-of-the-art methods in predicting in-hospital mortality and decompensation, demonstrating its effectiveness in reconstructing missing data and predicting patient health status. The code will be published at https://***/cling6666/SPIME. © 2024 IEEE.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Multi-Label Text Classification for Judicial Texts via Dual Graph and Label Feature Fusion 10

Multi-Label Text Classification for Judicial Texts via Dual ...

引用

10th IEEE Smart World Congress, SWC 2024

作者： Gu, Qiliang Lu, Qin Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Jinan China Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Jinan China Shandong Fundamental Research Center for Computer Science Shandong Provincial Key Laboratory of Industrial Network and Information System Security Jinan China

ISBN: (纸本)9798331520861

The legal judgement prediction (LJP) of judicial texts represents a multi-label text classification (MLTC) problem, which in turn involves three distinct tasks: the prediction of charges, legal articles, and terms of penalty. Nevertheless, extant multi-label text classification models tend to eschew the consideration of multiple correlations and semantic information between labels when predicting judicial texts, which may result in the loss of pertinent information. Furthermore, these models do not take full advantage of the local and overall information of the labels in the process of selecting appropriate labels for the text using the multi-head attention mechanism. To address these issues, we propose a new model, BGFLFF, which explores label correlation and semantics among the three tasks by employing Graph Convolutional network (GCN) and multi-head attention mechanisms. In particular, we propose a Bi-Graph Fusion GCN (BGF-GCN), which fuses the co-occurrence matrix of labels with the cosine similarity matrix, thereby fully exploiting the multiple correlations between labels. Furthermore, prior to embedding the tags, the interpretations of the labels are obtained via Google in order to enhance the semantic information between the labels. To further reinforce the interconnections between labels and text and more effectively comprehend the interrelationships between the three tasks, we propose a Multi-Head Label Feature Fusion Attentional Mechanism (MH-LFFAtt), which assigns distinct weights to the pertinent label information in the text by fusing the local and overall features of the labels. The experimental results demonstrate that the F1 score exhibits an improvement of up to 2. 2 8 % and a minimum of 1. 2 9 % across the various tasks of the two datasets. This evidence substantiates the assertion that BGFLFF outperforms the existing baseline model. © 2024 IEEE.

关键词： Semantics

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：