检索结果-内蒙古大学图书馆

Deep learning-based software engineering: progress,challenges, and opportunities

学校读者我要写书评

暂无评论

science China(Information sciences) 2025年第1期68卷 57-144页

作者： Xiangping CHEN Xing HU Yuan HUANG He JIANG Weixing JI Yanjie JIANG Yanyan JIANG Bo LIU Hui LIU Xiaochen LI Xiaoli LIAN Guozhu MENG Xin PENG Hailong SUN Lin SHI Bo WANG Chong WANG Jiayi WANG Tiantian WANG Jifeng XUAN Xin XIA Yibiao YANG Yixin YANG Li ZHANG Yuming ZHOU Lu ZHANG School of Journalism and Communication Sun Yat-sen University School of Software Technology Zhejiang University School of Software Engineering Sun Yat-sen University School of Software Dalian University of Technology School of Computer Science and Technology Beijing Institute of Technology Key Laboratory of High Confidence Software Technologies (Peking University) Ministry of EducationSchool of Computer Science Peking University State Key Laboratory for Novel Software Technology Nanjing University School of Computer Science and Engineering Beihang University Institute of Information Engineering Chinese Academy of Sciences School of Computer Science Fudan University State Key Laboratory of Complex & Critical Software Environment (CCSE) School of Software Beihang University School of Computer and Information Technology Beijing Jiaotong University School of Computer Science and Technology Harbin Institute of Technology School of Computer Science Wuhan University Huawei Technologies

Researchers have recently achieved significant advances in deep learning techniques, which in turn has substantially advanced other research disciplines, such as natural language processing, image processing, speech recognition, and software engineering. Various deep learning techniques have been successfully employed to facilitate software engineering tasks, including code generation, software refactoring, and fault localization. Many studies have also been presented in top conferences and journals, demonstrating the applications of deep learning techniques in resolving various software engineering tasks. However,although several surveys have provided overall pictures of the application of deep learning techniques in software engineering,they focus more on learning techniques, that is, what kind of deep learning techniques are employed and how deep models are trained or fine-tuned for software engineering tasks. We still lack surveys explaining the advances of subareas in software engineering driven by deep learning techniques, as well as challenges and opportunities in each subarea. To this end, in this study, we present the first task-oriented survey on deep learning-based software engineering. It covers twelve major software engineering subareas significantly impacted by deep learning techniques. Such subareas spread out through the whole lifecycle of software development and maintenance, including requirements engineering, software development, testing, maintenance, and developer collaboration. As we believe that deep learning may provide an opportunity to revolutionize the whole discipline of software engineering, providing one survey covering as many subareas as possible in software engineering can help future research push forward the frontier of deep learning-based software engineering more systematically. For each of the selected subareas,we highlight the major advances achieved by applying deep learning techniques with pointers to the available datasets i

关键词： deep learning software engineering software benchmark software artifact representation survey

Optimizing local search-based partial MaxSAT solving via initial assignment prediction

学校读者我要写书评

暂无评论

science China(Information sciences) 2025年第2期68卷 59-73页

作者： Chanjuan LIU Guangyuan LIU Chuan LUO Shaowei CAI Zhendong LEI Wenjie ZHANG Yi CHU Guojing ZHANG School of Computer Science and Technology Dalian University of Technology School of Software Beihang University Institute of Software Chinese Academy of Sciences School of Computer Science Peking University

Partial maximum satisfiability(PMS) is a significant generalization of Boolean satisfiability(SAT) and maximum satisfiability(MaxSAT), by introducing hard clauses and soft clauses. Compared with SAT and MaxSAT, the PMS problem has more real-world applications where both hard and soft constraints are involved. Local search is an effective incomplete method for solving PMS and is useful for important domains where good-quality solutions are desired within reasonable *** local search PMS solvers, the approach for initial assignment generation is crucial because its effectiveness significantly affects practical performance. In this study, we propose a novel initial assignment prediction approach, called InitPMS. When predicting an assignment for PMS, InitPMS considers the specific structure of PMS instances, i.e., distinguishing hard and soft clauses. Our experiments on extensive PMS instances from MaxSAT evaluations(MSEs) 2020 and 2021 show that InitPMS significantly boosts the performance of five state-of-the-art local search PMS solvers, demonstrating its generality. In addition,our results indicate that incorporating InitPMS could improve the performance of one of the best incomplete PMS solvers in MaxSAT Evaluation 2021, indicating that InitPMS might help advance the state of the art in PMS solving.

关键词： MaxSAT local search initial assignment incomplete method assignment prediction

Multi-Scale Feature Optimization Point Cloud Completion Network Integrating SoftPool

学校读者我要写书评

暂无评论

IAENG International Journal of computer science 2025年第1期52卷 233-243页

作者： Zhang, Wanpeng Zhou, Ziwei School of Computer Science and Software Engineering University of Science and Technology LiaoNing Anshan114051 China School of Computer Science and Software Engineering University of Science and Technology LiaoNing Anshan114051 China

Point cloud completion is crucial in point cloud processing, as it can repair and refine incomplete 3D data, ensuring more accurate models. However, current point cloud completion methods commonly face a challenge: they fail to fully utilize multi-scale information from local features, leading to limitations in accuracy and detail preservation. To address this issue, this paper proposes a multi-scale feature optimization algorithm for point cloud completion that integrates SoftPool. Based on DGCNN, the method combines dilated convolution and bottleneck attention mechanisms to extract features at different scales, enhancing the ability to capture detailed information in point clouds. The bottleneck attention mechanism is used to optimize important detail features. The extracted local features are concatenated with their corresponding positional information to form point proxies, enhancing the effective extraction of local geometric features, resulting in more refined completed point cloud shapes. A Transformer architecture is employed to model these features. Finally, SoftPool is introduced for fine-grained feature downsampling, improving the network's ability to recover point cloud details. FoldingNet is used to reconstruct missing structures and output the completed point cloud. To validate the model's completion performance, training and testing are conducted on the PCN and ShapeNet55 datasets. Experimental results demonstrate that the model has better feature detail retention and more accurate completion results. On the PCN dataset, the average CD value is reduced by 6.5% compared to the best-performing model among the comparison methods. On the ShapeNet55 dataset, the average CD value across three difficulty levels is reduced by 6.9% compared to the best-performing model among the comparison methods. Additionally, the model also achieved a 2.1% improvement in F-score. © (2025), (International Association of Engineers). All rights reserved.

关键词： Ability testing

AsyCo: an asymmetric dual-task co-training model for partial-label learning

学校读者我要写书评

暂无评论

science China(Information sciences) 2025年第5期68卷 332-347页

作者： Beibei LI Yiyuan ZHENG Beihong JIN Tao XIANG Haobo WANG Lei FENG College of Computer Science Chongqing University State Key Laboratory of Computer Science Institute of Software Chinese Academy of Sciences University of Chinese Academy of Sciences School of Software Technology Zhejiang University School of Computer Science and Engineering Nanyang Technological University

Partial-label learning(PLL) is a typical problem of weakly supervised learning, where each training instance is annotated with a set of candidate labels. Self-training PLL models achieve state-of-the-art performance but suffer from error accumulation problems caused by mistakenly disambiguated instances. Although co-training can alleviate this issue by training two networks simultaneously and allowing them to interact with each other, most existing co-training methods train two structurally identical networks with the same task, i.e., are symmetric, rendering it insufficient for them to correct each other due to their similar limitations. Therefore, in this paper, we propose an asymmetric dual-task co-training PLL model called AsyCo,which forces its two networks, i.e., a disambiguation network and an auxiliary network, to learn from different views explicitly by optimizing distinct tasks. Specifically, the disambiguation network is trained with a self-training PLL task to learn label confidence, while the auxiliary network is trained in a supervised learning paradigm to learn from the noisy pairwise similarity labels that are constructed according to the learned label confidence. Finally, the error accumulation problem is mitigated via information distillation and confidence refinement. Extensive experiments on both uniform and instance-dependent partially labeled datasets demonstrate the effectiveness of AsyCo.

关键词： machine learning weakly supervised learning partial-label learning co-training models candidate label sets

Improving UAV Image Target Detection: A Novel Approach Using OptiDETR with Swin Transformer

学校读者我要写书评

暂无评论

IAENG International Journal of computer science 2025年第3期52卷 771-780页

作者： Ma, Wenlong Liu, Weisheng School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China

In the analysis of drone aerial images, object detection tasks are particularly challenging, especially in the presence of complex terrain structures, extreme differences in target sizes, suboptimal shooting angles, and varying lighting conditions, all of which exacerbate the difficulty of recognition. In recent years, the DETR model based on the Transformer architecture has eliminated traditional post-processing steps such as NMS(Non-Maximum Suppression), thereby simplifying the object detection process and improving detection accuracy, which has garnered widespread attention in the academic community. However, DETR has limitations such as slow training convergence, difficulty in query optimization, and high computational costs, which hinder its application in practical fields. To address these issues, this paper proposes a new object detection model called OptiDETR. This model first employs a more efficient hybrid encoder to replace the traditional Transformer encoder. The new encoder significantly enhances feature processing capabilities through internal and cross-scale feature interaction and fusion logic. Secondly, an IoU (Intersection over Union) aware query selection mechanism is introduced. This mechanism adds IoU constraints during the training phase to provide higher-quality initial object queries for the decoder, significantly improving the decoding performance. Additionally, the OptiDETR model integrates SW-Block into the DETR decoder, leveraging the advantages of Swin Transformer in global context modeling and feature representation to further enhance the performance and efficiency of object detection. To tackle the problem of small object detection, this study innovatively employs the SAHI algorithm for data augmentation. Through a series of experiments, It achieved a significant performance improvement of more than two percentage points in the mAP (mean Average Precision) metric compared to current mainstream object detection models. Furthermore, ther

关键词： Decoding

Formal Verification of Data Modifications in Cloud Block Storage Based on Separation Logic

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2024年第1期33卷 112-127页

作者： Bowen ZHANG Zhao JIN Hanpin WANG Yongzhi CAO Key Laboratory of High Confidence Software Technologies (MOE) School of Computer Science Peking University School of Computer and Artificial Intelligence Zhengzhou University School of Computer Science and Cyber Engineering Guangzhou University

Cloud storage is now widely used, but its reliability has always been a major concern. Cloud block storage(CBS) is a famous type of cloud storage. It has the closest architecture to the underlying storage and can provide interfaces for other types. Data modifications in CBS have potential risks such as null reference or data *** verification of these operations can improve the reliability of CBS to some extent. Although separation logic is a mainstream approach to verifying program correctness, the complex architecture of CBS creates some challenges for verifications. This paper develops a proof system based on separation logic for verifying the CBS data modifications. The proof system can represent the CBS architecture, describe the properties of the CBS system state, and specify the behavior of CBS data modifications. Using the interactive verification approach from Coq, the proof system is implemented as a verification tool. With this tool, the paper builds machine-checked proofs for the functional correctness of CBS data modifications. This work can thus analyze the reliability of cloud storage from a formal perspective.

关键词： Formal verification Separation logic Cloud block storage Verification tool Coq

Self-supervised Hypergraph Transformer with Alignment and Uniformity for Recommendation

学校读者我要写书评

暂无评论

IAENG International Journal of computer science 2024年第3期51卷 292-300页

作者： Yang, XianFeng Liu, Yang School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan China School of computer Science and Software Engineering University of Science and Technology Liaoning Anshan China

Graph neural networks have proven their effectiveness for user-item interaction graph collaborative filtering. However, most of the existing recommendation models highly depended on abundant and high-quality datasets and neglected the alignment and uniformity of embedding representation. The noisy and skewed distribution of data from real-world applications is ubiquitous. The alignment and uniformity are crucial for representation learning. In this paper, we propose SHTAU (self-supervised hypergraph transformer with alignment and uniformity for the recommendation) to address the problem. Specifically, we first perform a graph neural network on the user-item interaction graph, then captures the global collaboration relationship between users and items through the hypergraph transformer structure, and use the extracted global information to generate self-supervised signals for data enhancement on the user-item interaction graph to enhance the robustness of the recommendation algorithm. Simultaneously, alignment and uniformity auxiliary task update embeddings from different angles to improve performance. Extensive experiments are conducted on Yelp, Gowalla, MovieLens, Amazon-book and Tmall datasets. The experimental results show that SHTAU has obvious advantages over the baseline methods. © (2024), (International Association of Engineers). All Rights Reserved.

关键词： Alignment

Surface Defect Detection Algorithm of Hot-Rolled Strip Based on Improved YOLOv7

学校读者我要写书评

暂无评论

IAENG International Journal of computer science 2024年第4期51卷 345-354页

作者： Shen, Lijia Cui, Wenhua Tao, Ye Shi, Tianwei Liao, Jinzhen School of Electronic and Information Engineering University of Science and Technology Liaoning Anshan China School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan China School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan China School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan China

To enhance the capability of classifying and localizing defects on the surface of hot-rolled strips, this paper proposed an algorithm based on YOLOv7 to improve defect detection. The BI-SPPFCSPC structure was incorporated into the feature pyramid in this algorithm, enabling enhanced extraction of features from small objects and improved accuracy in network model positioning. Additionally, a small object detection layer was introduced to enhance shallow feature capture. Then the CARAFE sampling operator was used for up-sampling to reduce the feature loss problem of small objects. Finally, the WIoU served as the loss function for network model training to expedite convergence. The NEU-DET dataset was utilized for ablation and comparison tests. The findings indicated that the enhanced YOLOv7 model's mAP value had increased to 80.7%. The detection impact was much enhanced in comparison to other traditional models, and the frequency of false and missing detections was also decreased. © (2024), (International Association of Engineers). All Rights Reserved.

关键词： Surface defects

Deep learning for code generation: a survey

学校读者我要写书评

暂无评论

science China(Information sciences) 2024年第9期67卷 5-40页

作者： Huangzhao ZHANG Kechi ZHANG Zhuo LI Jia LI Jia LI Yongmin LI Yunfei ZHAO Yuqi ZHU Fang LIU Ge LI Zhi JIN Key Lab of High Confidence Software Technologies (Peking University) Ministry of Education School of Computer Science Peking University School of Computer Science and Engineering Beihang University

In the past decade, thanks to the powerfulness of deep-learning techniques, we have witnessed a whole new era of automated code generation. To sort out developments, we have conducted a comprehensive review of solutions to deep learning-based code generation. In this survey, we generally formalize the pipeline and procedure of code generation and categorize existing solutions according to taxonomy from perspectives of architecture, model-agnostic enhancing strategy, metrics, and tasks. In addition, we outline the challenges faced by current dominant large models and list several plausible directions for future research. We hope that this survey may provide handy guidance to understanding, utilizing, and developing deep learning-based code-generation techniques for researchers and practitioners.

关键词： code generation automated software engineering deep learning large model artificial intelligence