检索结果-内蒙古大学图书馆

Block-gram:Mining knowledgeable features for efficiently smart contract vulnerability detection

Digital Communications and Networks 2025年第1期11卷 1-12页

作者： Xueshuo Xie Haolong Wang Zhaolong Jian Yaozheng Fang Zichun Wang Tao Li Tianjin Key Laboratory of Network and Data Security Technology TianjinChina College of Computer Science Nankai UniversityTianjinChina Key Laboratory of Blockchain and Cyberspace Governance of Zhejiang Province China State Key Laboratory of Computer Architecture Institute of Computing TechnologyChinese Academy of SciencesChina

Smart contracts are widely used on the blockchain to implement complex transactions,such as decentralized applications on *** vulnerability detection of large-scale smart contracts is critical,as attacks on smart contracts often cause huge economic *** it is difficult to repair and update smart contracts,it is necessary to find the vulnerabilities before they are ***,code analysis,which requires traversal paths,and learning methods,which require many features to be trained,are too time-consuming to detect large-scale on-chain ***-based methods will obtain detection models from a feature space compared to code analysis methods such as symbol *** the existing features lack the interpretability of the detection results and training model,even worse,the large-scale feature space also affects the efficiency of *** paper focuses on improving the detection efficiency by reducing the dimension of the features,combined with expert *** this paper,a feature extraction model Block-gram is proposed to form low-dimensional knowledge-based features from ***,the metadata is separated and the runtime code is converted into a sequence of opcodes,which are divided into segments based on some instructions(jumps,etc.).Then,scalable Block-gram features,including 4-dimensional block features and 8-dimensional attribute features,are mined for the learning-based model ***,feature contributions are calculated from SHAP values to measure the relationship between our features and the results of the detection *** addition,six types of vulnerability labels are made on a dataset containing 33,885 contracts,and these knowledge-based features are evaluated using seven state-of-the-art learning algorithms,which show that the average detection latency speeds up 25×to 650×,compared with the features extracted by N-gram,and also can enhance the interpretability of the detection model.

关键词： Smart contract Bytecode&opcode Knowledgeable features Vulnerability detection Feature contribution

来源：评论

学校读者我要写书评

暂无评论

Joint User Association,Resource Allocation and Trajectory Design for Multi-UAV-Aided NOMA Wireless Communication Systems

引用

China Communications 2025年第3期22卷 217-233页

作者： Yin Sixing Qu Zhaowei Yu Peng Beijing Key Laboratory of Network System Architecture and Convergence Beijing University of Posts and TelecommunicationsBeijing 100876China School of Computer Science Beijing University of Posts and TelecommunicationsBeijing 100876China

In this paper,we investigate a multi-UAV aided NOMA communication system,where multiple UAV-mounted aerial base stations are employed to serve ground users in the downlink NOMA communication,and each UAV serves its associated users on its own *** aim at maximizing the overall common throughput in a finite time *** a problem is a typical mixed integer nonlinear problem,which involves both continuous-variable and combinatorial *** efficiently solve this problem,we propose a two-layer algorithm,which separately tackles continuous-variable and combinatorial ***,in the inner layer given one user association scheme,subproblems of bandwidth allocation,power allocation and trajectory design are solved based on alternating *** the outer layer,a small number of candidate user association schemes are generated from an initial scheme and the best solution can be determined by comparing all the candidate *** particular,a clustering algorithm based on K-means is applied to produce all candidate user association schemes,the successive convex optimization technique is adopted in the power allocation subproblem and a logistic function approximation approach is employed in the trajectory design *** results show that the proposed NOMA scheme outperforms three baseline schemes in downlink common throughput,including one solution proposed in an existing literature.

关键词： NOMA resource allocation trajectory design UAV communications user association

来源：评论

学校读者我要写书评

暂无评论

Path Optimization Method Under UAV Charging Scheduling Network 24th

Path Optimization Method Under UAV Charging Scheduling Netwo...

引用

24th International Conference on Algorithms and architectures for Parallel Processing, ICA3PP 2024

作者： Yang, Tingting Wang, Yiqian Zhu, Jie Chang, Shuyu Huang, Haiping Nanjing University of Posts and Telecommunications Nanjing210003 China State Key Laboratory Chinese of Computer Architecture Institute of Computing Technology Academy of Sciences Beijing100864 China Jiangsu High Technology Research Key Laboratory for Wireless Sensor Networks Nanjing210003 China

ISBN: (纸本)9789819615278

Unmanned Aerial Vehicles (UAVs) have emerged as integral components in logistics systems, where their potential for efficient delivery services is being explored. However, the limited battery capacity of UAVs poses a significant challenge for long-distance delivery applications. In this paper, we investigate a hybrid problem involving UAV trajectory planning and charging scheduling for long-distance logistics systems with a charging network. In the problem under study, multiple UAVs are assigned long-distance delivery tasks. A UAV can stop at any charging station in the charging network to charge if it is in a low-energy state. An Ant Colony Optimization-based UAV Trajectory Planning and Charging Scheduling (ACO-TPCS) algorithm is proposed to minimize task completion time by optimizing UAV trajectory and charging plans. The main framework of the ACO-TPCS algorithm consists of several key components, including reachable graph construction, candidate flight path generation, pheromone matrix construction, ant generation, feasible solution generation, and pheromone update method. All these components are delicately designed. Through extensive experimentation and comparison with baseline algorithms, we demonstrate the effectiveness of the ACO-TPCS algorithm in addressing the problem under study. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Ant colony optimization

来源：评论

学校读者我要写书评

暂无评论

Dual-step optimization for binary sequences with high merit factors

引用

Digital Signal Processing: A Review Journal 2025年 165卷

作者： Pšeničnik, Blaž Mlinarič, Rene Brest, Janez Bošković, Borko Computer Architecture and Languages Laboratory Institute of Computer Science Faculty of Electrical Engineering and Computer Science University of Maribor Koroška c. 46 Maribor2000 Slovenia

The problem of finding aperiodic low auto-correlation binary sequences (LABS) presents a significant computational challenge, particularly as the sequence length increases. Such sequences have important applications in communication engineering, physics, chemistry, and cryptography. This paper introduces a dual-step algorithm for long binary sequences with high merit factors. The first step employs a parallel algorithm utilizing skew-symmetry and restriction classes to generate sequence candidates with merit factors above a predefined threshold. The second step uses a priority queue algorithm to refine these candidates further, searching the entire search space unrestrictedly. By combining GPU-based parallel computing and dual-step optimization, our approach has successfully identified best-known binary sequences for all lengths ranging from 450 to 527, with the exception of length 518, where the previous best-known merit factor value was matched with a different sequence. This hybrid method significantly outperforms traditional exhaustive and stochastic search methods, offering an efficient solution for finding long sequences with good merit factors. © 2025 The Author(s)

关键词： Digital arithmetic

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Based Cellular Nuclei Segmentation Using Transformer Model 5th

Deep Learning-Based Cellular Nuclei Segmentation Using Trans...

引用

5th Workshop on Intelligent Data - From Data to Knowledge, DOING 2024, 3rd Workshop on Knowledge Graphs Analysis on a Large Scale, K-GALS 2024, 6th Workshop on Modern Approaches in Data Engineering and Information System Design, MADEISD 2024 and 3rd Workshop on Personalization and Recommender Systems, PERS 2024 held in conjunction with 28th European Conference on Advances in Databases and Information Systems, ADBIS 2024

作者： Erezman, Mateusz Dziubich, Tomasz Computer Vision and Artificial Intelligence Laboratory Department of Computer Architecture Faculty of Electronics Telecommunications and Informatics Gdańsk University of Technology Gdańsk Poland

ISBN: (纸本)9783031704208

Accurate segmentation of cellular nuclei is imperative for various biological and medical applications, such as cancer diagnosis and drug discovery. Histopathology, a discipline employing microscopic examination of bodily tissues, serves as a cornerstone for cancer diagnosis. Nonetheless, the conventional histopathological diagnosis process is frequently marred by time constraints and potential inaccuracies. Consequently, there arises a pressing need for automated image analysis tools to augment medical practitioners’ efforts. In this paper, we present a novel approach utilising Transformer model, originally designed for natural language processing tasks, for automated cellular nuclei segmentation in whole-slide microscopic images. Specifically targeting cell nuclei, this methodology holds significance as the initial phase in diagnosing various illnesses, streamlining the analysis and quantification process. The study introduces a novel model that combines a U-Net architecture with a Transformer-based network functioning as a parallel encoder. This model was compared against three other popular architectures in the literature: U-Net, ResU-Net, and LinkNet-34. The impact of augmentation and colour normalisation techniques was investigated. The average Dice similarity coefficient for the considered images was found to be 0.8041. The obtained results seem to be clinically relevant. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

HIET:Hybrid Information Enhancement Transformer Network for Single-Photon Image Reconstruction

引用

Journal of Beijing Institute of Technology 2025年第1期34卷 1-17页

作者： Yiming Liu Xuri Yao Tao Zhang Yifei Sun Ying Fu School of Computer Science and Technology Beijing Institute of TechnologyBeijing 100081China Center for Quantum Technology Research and Key Laboratory of Advanced Optoelec-tronic Quantum Architecture and Measurements(MOE) School of PhysicsBeijing Institute of TechnologyBeijing 100081China School of Communication Engineering Hangzhou Dianzi UniversityHangzhou 310018China Uniview Technology Hangzhou 310056China

Single-photon sensors are novel devices with extremely high single-photon sensitivity and temporal ***,these advantages also make them highly susceptible to ***,single-photon cameras face severe quantization as low as 1 bit/*** factors make it a daunting task to recover high-quality scene information from noisy single-photon *** current image reconstruction methods for single-photon data are mathematical approaches,which limits information utilization and algorithm *** this work,we propose a hybrid information enhancement model which can significantly enhance the efficiency of information utilization by leveraging attention mechanisms from both spatial and channel ***,we introduce a structural feature enhance module for the FFN of the transformer,which explicitly improves the model's ability to extract and enhance high-frequency structural information through two symmetric convolution ***,we propose a single-photon data simulation pipeline based on RAW images to address the challenge of the lack of single-photon *** results show that the proposed method outperforms state-of-the-art methods in various noise levels and exhibits a more efficient capability for recovering high-frequency structures and extracting information.

关键词： single-photon images hybrid information enhancement structual feature enhancement data simulation pipeline

来源：评论

学校读者我要写书评

暂无评论

Optimizing Self-training Sample Selection for Euphemism Detection in Special Scenarios 24th

Optimizing Self-training Sample Selection for Euphemism Dete...

引用

24th International Conference on Algorithms and architectures for Parallel Processing, ICA3PP 2024

作者： Zhang, Zhiwei Chang, Shuyu Huang, Haiping Wang, Rui Nanjing University of Posts and Telecommunications Nanjing210023 China Jiangsu High Technology Research Key Laboratory for Wireless Sensor Networks Nanjing210023 China Nanjing University of Posts and Telecommunications Nanjing210023 China State Key Laboratory Chinese of Computer Architecture Institute of Computing Technology Academy of Sciences Beijing100864 China

ISBN: (纸本)9789819615476

Euphemism, which is often used to disguise true intentions or soften the tone of speech, has garnered significant attention for its detection in recent years. Current approaches to euphemism detection struggle with uncommon data, where new and rare vocabularies frequently emerge in the test set, leading to decreased model performance. To address these challenges, we propose a novel framework that combines data augmentation, self-training, and a special sample selection method. We expand the original dataset using lexical and sentence-level data augmentation techniques, enhancing the representation of rare vocabulary and allowing the model to familiarize itself with their contexts and usage characteristics. Next, we employ self-training as a semi-supervised learning approach, utilizing the augmented datasets to iteratively generate and refine pseudo-labels. During each iteration, a KNN framework is used to generate pseudo-labels by searching for the nearest neighbor texts, thus enhancing the quality of these labels. Finally, to further increase the presence of rare vocabulary in the training data, we implement a sample selection strategy based on euphemism frequency and predictive uncertainty, integrating it with the earlier steps. Experimental results demonstrate that our framework outperforms previous research on most metrics and achieves a 10% increase over the baseline model. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Supply and Demand Relationship of Fresh Retail Based on Accessibility: A Case Study of Guangzhou

Supply and Demand Relationship of Fresh Retail Based on Acce...

引用

2024 International Conference on Construction and Real Estate Management: ESG Development in the Construction Industry, ICCREM 2024

作者： Jiang, Rui Hu, Zilin Wei, Zongcai Wu, Kaixin Liang, Jing Liu, Ziye School of Architecture South China Univ. of Technology Guangzhou China State Key Laboratory of Subtropical Building and Urban Science South China Univ.of Technology Guangzhou China School of Computer Science and Engineering South China Univ. of Technology Guangzhou China

ISBN: (数字)9780784485910

ISBN: (纸本)9780784485910

Whether the supply-demand relationship of fresh food retail services matches is of great significance to urban planning and construction and residents' lives. This work takes four central districts in Guangzhou as an example to calculate the accessibility of fresh retail in different types of residential areas. In addition, this work explores the spatial clustering distribution relationship of fresh retail supply and demand. Results show that the frequency of purchasing fresh food in central areas is typically higher than that in border areas. However, the supply-demand matching is poor due to the large population size. The online accessibility of newly built urban areas, such as Tianhe District, is better, but offline fresh food retailing dominates. Offline fresh food retail supply and demand matching is better than online in the marginal area. The analysis of the accessibility of fresh food in residential areas is beneficial for providing a reference for urban spatial layout and planning and construction. © ASCE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Fresh Retail Layout and Forecast Based on Multi-Source Big Data and Machine Learning: A Case Study of Guangzhou

Fresh Retail Layout and Forecast Based on Multi-Source Big D...

引用

2024 International Conference on Construction and Real Estate Management: ESG Development in the Construction Industry, ICCREM 2024

作者： Wu, Kaixin Liu, Ziye Wei, Zongcai Liang, Jing Jiang, Rui Hu, Zilin School of Architecture South China Univ. of Technology Guangzhou China School of Computer Science and Engineering South China Univ. of Technology Guangzhou China State Key Laboratory of Subtropical Building and Urban Science South China Univ. of Technology Guangzhou China

ISBN: (数字)9780784485910

ISBN: (纸本)9780784485910

Fresh retail is an important facility in the community life circle. Exploring its spatial distribution is essential for improving the allocation of life service resources. This paper introduces a method based on supply and demand matching and machine learning, designed to analyze and forecast the current state of supply and demand in offline fresh retail. Specifically, we take the four central districts of Guangzhou as the research object and apply the two-step mobile search method to filter the training set grid. Utilizing machine learning, we obtained the feature matrices of these grids and applied them to the supply prediction of the remaining grids. The results indicate that the Commercial Residential Density, Travel Convenience Index, and Food Service Density are the main factors affecting the location selection of offline fresh retail. The supply and demand in Yuexiu and Haizhu Districts show imbalance, and the construction of offline fresh retail facilities must be fortified. © ASCE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

RGB-Based Category-Level Object Pose Estimation via Depth Recovery and Adaptive Refinement

引用

IEEE Robotics and Automation Letters 2025年第6期10卷 5377-5384页

作者： Yang, Hui Sun, Wei Liu, Jian Zheng, Jin Zeng, Zhiwen Mian, Ajmal Hunan University National Engineering Research Center for Robot Visual Perception and Control Technology College of Electrical and Information Engineering Changsha410082 China Hunan University State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body Changsha410082 China Central South University School of Architecture and Art Changsha410082 China University of Western Australia Department of Computer Science and Software Engineering CrawleyWA6009 Australia

Category-level pose estimation methods have received widespread attention as they can be generalized to intra-class unseen objects. Although RGB-D-based category-level methods have made significant progress, reliance on depth image limits practical application. RGB-based methods offer a more practical and cost-effective solution. However, current RGB-based methods struggle with object geometry perception, leading to inaccurate pose estimation. We propose depth recovery and adaptive refinement for category-level object pose estimation from a single RGB image. We leverage DINOv2 to reconstruct the coarse scene-level depth from the input RGB image and propose an adaptive refinement network based on an encoder-decoder architecture to dynamically improve the predicted coarse depth and reduce its gap from the ground truth. We introduce a 2D–3D consistency loss to ensure correspondence between the point cloud obtained from depth projection and the objects in the 2D image. This consistency supervision enables the model to maintain alignment between the depth image and the point cloud. Finally, we extract features from the refined point cloud and feed them into two confidence-aware rotation regression branches and a translation and size prediction residual branch for end-to-end training. Decoupling the rotation matrix provides a more direct representation, which facilitates parameter optimization and gradient propagation. Extensive experiments on the REAL275 and CAMERA25 datasets demonstrate the superior performance of our method. Real-world estimation and robotic grasping experiments demonstrate our model robustness to occlusion, clutter environments, and low-textured objects. Our code and robotic grasping video are available at DA-Pose. © 2016 IEEE.

关键词： Depth perception

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：