检索结果-内蒙古大学图书馆

LEGF-DST:LLMs-Enhanced Graph-Fusion Dual-Stream Transformer for Fine-Grained Chinese Malicious SMS Detection

computers, Materials & Continua 2025年第2期82卷 1901-1924页

作者： Xin Tong Jingya Wang Ying Yang Tian Peng Hanming Zhai Guangming Ling School of Information and Cybersecurity People’s Public Security University of ChinaBeijing100038China Cyber Investigation Technology Research and Development Center The Third Research Institute of the Ministry of Public SecurityShanghai201204China Department of Cybersecurity Defense Beijing Police CollegeBeijing102202China School of Computer Science Henan Institute of EngineeringZhengzhou451191China

With the widespread use of SMS(Short Message Service),the proliferation of malicious SMS has emerged as a pressing societal *** deep learning-based text classifiers offer promise,they often exhibit suboptimal performance in fine-grained detection tasks,primarily due to imbalanced datasets and insufficient model representation *** address this challenge,this paper proposes an LLMs-enhanced graph fusion dual-stream Transformer model for fine-grained Chinese malicious SMS *** the data processing stage,Large Language Models(LLMs)are employed for data augmentation,mitigating dataset *** the data input stage,both word-level and character-level features are utilized as model inputs,enhancing the richness of features and preventing information loss.A dual-stream Transformer serves as the backbone network in the learning representation stage,complemented by a graph-based feature fusion *** the output stage,both supervised classification cross-entropy loss and supervised contrastive learning loss are used as multi-task optimization objectives,further enhancing the model’s feature *** results demonstrate that the proposed method significantly outperforms baselines on a publicly available Chinese malicious SMS dataset.

关键词： Transformers malicious SMS multi-task learning large language models

来源：评论

学校读者我要写书评

暂无评论

BTM: Black-Box Testing for DNN Based on Meta-Learning 23

BTM: Black-Box Testing for DNN Based on Meta-Learning

引用

23rd IEEE International Conference on software Quality, Reliability, and Security, QRS 2023

作者： Zhang, Zhuangyu Zhang, Zhiyi Wang, Ziyuan Chen, Fang Huang, Zhiqiu Nanjing University of Aeronautics and Astronautics Collage of Computer Science and Technology Nanjing China Collaborative Innovation Center of Novel Software Technology and Industrialization Nanjing China Nanjing University of Posts and Telecommunications School of Computer Science and Technology Nanjing China Nanjing University of Aeronautics and Astronautics Ministry Key Laboratory for Safety-Critical Software Development and Verification Nanjing China

ISBN: (纸本)9798350319583

Deep learning is widely used in security fields like autonomous driving, but testing deep learning models poses challenges due to low generation efficiency and limited error detection. Current white-box test case generation methods rely on neuron coverage, but black-box testing is crucial when model details cannot be accessed. Prior methods also needed more consideration for error diversity under lower time resources. In this paper, we propose a novel approach based on the intuition that two models trained on the same classification task learn similar rules at a coarse-grained level. We employ meta-learning to train universal meta perturbation on a surrogate model, which increases neuron coverage on the surrogate model while inducing misclassifications in the original model (thus enhancing the adequacy of the initial tests). Subsequently, we apply data initialization to discover a wider range of errors faster. To accelerate test case generation and reduce resource consumption, we introduce a set of acceleration techniques based on image prediction probabilities, minimizing the time spent exploring irrelevant regions in images. Finally, we evaluate our approach on a well-known dataset and several well-known models, demonstrating its effectiveness. © 2023 IEEE.

关键词： Black-box testing

来源：评论

学校读者我要写书评

暂无评论

Real-IAD: A Real-World Multi-View Dataset for Benchmarking Versatile Industrial Anomaly Detection

arXiv

引用

arXiv 2024年

作者： Wang, Chengjie Zhu, Wenbing Gao, Bin-Bin Gan, Zhenye Zhang, Jiangning Gu, Zhihao Qian, Shuguang Chen, Mingang Ma, Lizhuang Shanghai Jiao Tong University China Youtu Lab Tencent China Fudan University China Rongcheer Co. Ltd China Shanghai Development Center of Computer Software Technology China

Industrial anomaly detection (IAD) has garnered significant attention and experienced rapid development. However, the recent development of IAD approach has encountered certain difficulties due to dataset limitations. On the one hand, most of the state-of-the-art methods have achieved saturation (over 99% in AUROC) on mainstream datasets such as MVTec, and the differences of methods cannot be well distinguished, leading to a significant gap between public datasets and actual application scenarios. On the other hand, the research on various new practical anomaly detection settings is limited by the scale of the dataset, posing a risk of overfitting in evaluation results. Therefore, we propose a large-scale, Real-world, and multi-view Industrial Anomaly Detection dataset, named Real-IAD, which contains 150K high-resolution images of 30 different objects, an order of magnitude larger than existing datasets. It has a larger range of defect area and ratio proportions, making it more challenging than previous datasets. To make the dataset closer to real application scenarios, we adopted a multi-view shooting method and proposed sample-level evaluation metrics. In addition, beyond the general unsupervised anomaly detection setting, we propose a new setting for Fully Unsupervised Industrial Anomaly Detection (FUIAD) based on the observation that the yield rate in industrial production is usually greater than 60%, which has more practical application value. Finally, we report the results of popular IAD methods on the Real-IAD dataset, providing a highly challenging benchmark to promote the development of the IAD field. Copyright © 2024, The Authors. All rights reserved.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Prototype Learning Based Realistic 3D Terrain Generation from User Semantics 22nd

Prototype Learning Based Realistic 3D Terrain Generation fro...

引用

22nd Asia Simulation Conference, AsiaSim 2023

作者： Gao, Yan Li, Jimeng Xu, Jianzhong Song, Xiao Quan, Hongyan School of Computer Science and Technology East China Normal University Shanghai200062 China School of Software Engineering East China Normal University Shanghai200062 China Training and Simulation Center Army Infantry Academy Jiangxi Nanchang330103 China School of Cyber Science and Technology Beihang University Beijing100191 China

ISBN: (纸本)9789819972395

Customizing 3D terrain based on user semantics plays an important role in military simulation, but it is difficult to realize realistic results because of the limited ability of some simple Convolutional neural network (CNN) models. In order to meet the personalized needs of users, this article proposes a prototype learning based terrain generation network (ProTG Net). Concretely, it extracts terrain semantics based prototype features from a small number of terrain surface samples, and then transfers the pre-learned features to user customization. Specifically, a prototype learning based framework is designed, including a terrain texture generation module (TGM), prototype feature generation module (PGM), and multiple prototype features matching module (FMM). TGM is designed as the CGAN based Pix2pix (Pixel to Pixel) structure, which can generate realistic terrain textures based on user semantics, providing a reliable terrain texture data source for prototype learning. Based on the semantic terrain texture generated by TGM, multi-features are extracted in PGM including the adaptive super-pixel guided features and the terrain spatial feature. In addition, multiple feature matching strategy is proposed for achieving the better matching between prototype matching and user semantic features. Taking a public dataset of real terrain as an example, it was verified that the prototype based method can generate realistic 3D terrain and achieve user customization to obtain realistic results. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Rectal Cancer Stages T2 and T3 Identification Based on Asymptotic Hybrid Feature Maps

引用

computer Modeling in Engineering & Sciences 2023年第10期137卷 923-938页

作者： Shujing Sun Jiale Wu Jian Yao Yang Cheng Xin Zhang Zhihua Lu Pengjiang Qian R&D Center of Artificial Intelligence Systems and Applications School of Artificial Intelligence and Computer ScienceJiangnan UniversityWuxi214122China Shanghai Enterprise Information Operation Center China Telecom Group Co.Ltd.Shanghai201315China Department of Radiology Dushu Lake Hospital Affiliated to Soochow UniversitySuzhou215123China Suzhou Yunmai Software Technology Co. Ltd.Suzhou215100China

Many existing intelligent recognition technologies require huge datasets for model ***,it is not easy to collect rectal cancer images,so the performance is usually low with limited training *** addition,traditional rectal cancer staging is time-consuming,error-prone,and susceptible to physicians’subjective awareness as well as professional *** settle these deficiencies,we propose a novel deep-learning model to classify the rectal cancer stages of T2 and ***,a novel deep learning model(RectalNet)is constructed based on residual learning,which combines the squeeze-excitation with the asymptotic output layer and new cross-convolution layer links in the residual block ***,a two-stage data augmentation is designed to increase the number of images and reduce deep learning’s dependence on the volume of *** experiment results demonstrate that the proposed method is superior to many existing ones,with an overall accuracy of ***,other traditional techniques,such as VGG16,DenseNet121,EL,and DERNet,have an average accuracy of 0.6981,0.7032,0.7500,and 0.7685,respectively.

关键词： Rectal cancer staging T2WI images data expansion asymptotic output layer

来源：评论

学校读者我要写书评

暂无评论

Method of adaptive learning feature model fusion and verification based on metamodels 3

Method of adaptive learning feature model fusion and verific...

引用

3rd International Symposium on computer Applications and Information Systems, ISCAIS 2024

作者： Wang, Chenyu Zhu, Yi Wu, Zixin Hao, Guosheng School of Computer Science and Technology Jiangsu Normal University Xuzhou221000 China Jiangsu Province Education Informatization Engineering Technology Research Center Jiangsu Normal University Xuzhou221000 China The Key Laboratory of the Software Development and Verification Technology The Ministry of Industry and Information Technology Nanjing University of Aeronautics and Astronautics Nanjing210016 China

ISBN: (纸本)9781510681316

As the advancement of computer technique and artificial intelligence technology, education based on computer technology has become a hot topic, and the wave of educational informatization is rising. Among them, adaptive learning has received widespread attention as a type of educational informatization technology. However, the current adaptive learning feature models use different semantics to describe each other, and the models lack meta model level explanations, resulting in a lack of further operability of the models and difficulty in integrating with each other. In order to solve these problems, our paper proposes a fusion and verification method for learning feature adaptive models based on meta models. Above all, an adaptive learning feature meta model is proposed, which defines the construction rules of the adaptive learning feature model at meta model level, so that different models can be fused using the same set of semantic rules. Then, at the meta model level, transform the learning model into a requirement model. Finally, using formal methods, validate the fused feature model, which can make sure of the semantic consistency between the previous and subsequent models. This article demonstrates the entire process through a model fusion experiment. The experimental results indicate that this method is correct and feasible in promoting feature model fusion without adaptive learning. © 2024 SPIE.

关键词： Informatization

来源：评论

学校读者我要写书评

暂无评论

Research on Recognition Method of Floating Objects on Water Surface Based on improved ResNet 6

Research on Recognition Method of Floating Objects on Water ...

引用

6th International Conference on Electronic Information technology and computer Engineering, EITCE 2022

作者： Guo, Haonan Ma, Yangyi Hua, Lei Wan, D. College of Computer and Information Hohai University China Software Development Center Bank of Communications China School of Management Tianjin University of Technology China

ISBN: (纸本)9781450397148

With the continuous population growth and rapid economic development, problems such as scarcity and pollution of freshwater resources are increasingly emerging. Floating objects on the water surface are one of the causes of water pollution, so the research on the identification algorithm of floating objects is of great significance. To further improve the accuracy of the identification of floating objects on the water surface, this paper proposes a DS-ResNet model based on the improved ResNet. Compared to previous work, DS-ResNet employs several innovations to improve training and testing speed while also increasing recognition accuracy. We established an actual data sample set in the field of rivers and lakes, and a modified ResNet model was constructed by Tensorflow for training. The experimental results show that the training cycle time of DS-ResNet is reduced by about 7% compared with the traditional ResNet, and the recognition accuracy of common targets reaches 98%. © 2022 Association for Computing Machinery.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Blocking Flow Shop Scheduling Based on a Multi-Group Collaborative Sparrow Search Algorithm

Blocking Flow Shop Scheduling Based on a Multi-Group Collabo...

引用

第34届中国控制与决策会议

作者： Kai Ma Yafei Geng Kan Xu Shiliang Guo Jie Yang School of Electrical Engineering Yanshan University Shanghai Development Center of Computer Software Technology

The blocking flow shop scheduling problem(BFSP) prevails widely in the process manufacturing *** this paper,a multi-group collaborative sparrow search algorithm(MGCSSA) is proposed for minimizing the makespan for ***,a collaborative initialization strategy is designed to generate initialized populations with certain ***,an elite solution set and a multi-group collaboration mechanism are introduced to improve the adequacy of the global search,and the producer and scrounger position update formulas are improved to increase the population ***,three neighborhood heuristics are used to locally strengthen individuals in the elite solution set that enhance the ability of the algorithm to jump out of the local ***,the proposed algorithm is evaluated experimentally in test *** experimental data and comparative results show that the proposed MGCSSA has superior performance and is highly competitive in solving BFSP.

关键词： Blocking Flow Shop Scheduling Hybrid Modified Global Optimal Sparrow Search Algorithm Neighborhood Heuristics

来源：评论

学校读者我要写书评

暂无评论

Joint Learning Content and Degradation Aware Feature for Blind Super-Resolution

arXiv

引用

arXiv 2022年

作者： Zhou, Yifeng Lin, Chuming Luo, Donghao Liu, Yong Tai, Ying Wang, Chengjie Chen, Mingang Youtu Lab Tencent China Shanghai Development Center of Computer Software Technology China

To achieve promising results on blind image super-resolution (SR), some attempts leveraged the low resolution (LR) images to predict the kernel and improve the SR performance. However, these Supervised Kernel Prediction (SKP) methods are impractical due to the unavailable real-world blur kernels. Although some Unsupervised Degradation Prediction (UDP) methods are proposed to bypass this problem, the inconsistency between degradation embedding and SR feature is still challenging. By exploring the correlations between degradation embedding and SR feature, we observe that jointly learning the content and degradation aware feature is optimal. Based on this observation, a Content and Degradation aware SR Network dubbed CDSR is proposed. Specifically, CDSR contains three newly-established modules: (1) a Lightweight Patch-based Encoder (LPE) is applied to jointly extract content and degradation features;(2) a Domain Query Attention based module (DQA) is employed to adaptively reduce the inconsistency;(3) a Codebook-based Space Compress module (CSC) that can suppress the redundant information. Extensive experiments on several benchmarks demonstrate that the proposed CDSR outperforms the existing UDP models and achieves competitive performance on PSNR and SSIM even compared with the state-of-the-art SKP methods. Copyright © 2022, The Authors. All rights reserved.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

center Region Aware Cross Former For Image Forgery Localization 15

Center Region Aware Cross Former For Image Forgery Localizat...

引用

15th International Conference on Signal Processing Systems, ICSPS 2023

作者： Wu, Yahui Gong, Xuchao Zhang, Miao Zhang, Peng SenseTime Inc Caohejing Emerging Technology Development 1900 Hongmei Road Zone Shanghai China Information Technology Service Center of Shengli Petroleum Management Bureau Shangdong Dongying China School of Computer Science and Information Engineering Hubei University Hubei Wuhan China

ISBN: (纸本)9781510675056

Image forgery has been a serious issue in real life during this boosting big data era. There have been many methods depending on detecting footprints such as edge inconsistency, camera noise, JPEG artifacts, etc., proposed. Unlike the existing methods, we propose a more general method, CRAC-formerNet to capture more stable manipulation traces, not only including RGB domain but also from high-frequency features generated from Steganalysis Rich Model (SRM)1 and District Cosine transformer in the frequency domain. A shared query is generated from features from both the RGB and frequency domains. The keys and values are generated from each domain respectively. Moreover, features of manipulated and manipulated regions should belong to different distributions, especially features in edge regions. We use a center contrastive loss to learn fine-grained features. A manipulated region proposal module is proposed for improving computing efficiency when calculating the loss. We empirically demonstrate that our method achieves competitive results on four benchmark image manipulation datasets. © 2024 SPIE.

关键词： Frequency domain analysis

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：