Information and Software Technology最新文献_第9页

Are cloud providers exploiting open-source? An exploratory study of Redis license change 云提供商是否在利用开源？Redis许可变更的探索性研究

IF 4.3 2区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information and Software Technology

Pub Date : 2026-02-01 Epub Date: 2025-11-03 DOI: 10.1016/j.infsof.2025.107951

Fanyu Han , Shengyu Zhao , Xiaoya Xia , Wei Wang

Context:

On March 20, 2024, Redis Inc. changed the Redis project’s license from the permissive Berkeley Software Distribution (BSD) license to a dual licensing model under the Redis Source Available License (RSALv2) and the Server Side Public License (SSPLv1). The official rationale for this change was to restrict cloud service providers from offering Redis as a managed service without contributing back. The license transition drew widespread attention within the open source community and raised concerns from some developers and organizations about its implications for project governance, contribution dynamics, and long-term sustainability.

Objective:

Analyze the contribution distribution within the Redis repository by examining the behavior of developers with different motivations, evaluate the changes that occurred during the license change period, and assess whether cloud providers are exploiting the open-source project.

Method:

This study categorizes developers’ motivations based on collaboration behavior data. By leveraging developer profile information, contributors are categorized into three motivation types: company-driven, community-driven, and communication-driven, enabling an analysis of trends across different contributor groups. The study primarily relies on project evaluation metrics from the Community Health Analytics Open Source Software (CHAOSS) community, along with collaboration metrics, to examine changes within the Redis repository.

Results:

Since 2017, cloud providers have consistently contributed to the Redis open-source project. During the license change period, there was a notable decrease in contributor participation and influence, particularly among company-driven and community-driven developers. Several core contributors transitioned to the newly established Valkey project.

Conclusion:

Following the license change by Redis Inc., the community experienced a certain degree of fragmentation, with major cloud providers migrating to the Valkey fork. Cloud providers have recognized the importance of the community and are willing to invest resources into the open-source projects they participate in, ensuring better collaboration and alignment with upstream development for their cloud services.

背景：在2024年3月20日，Redis公司将Redis项目的许可证从宽松的Berkeley Software Distribution （BSD）许可证更改为Redis Source Available license （RSALv2）和Server Side Public license （SSPLv1）下的双重许可模式。这一变化的官方理由是限制云服务提供商在没有回报的情况下将Redis作为托管服务提供。许可证的转换引起了开放源代码社区的广泛关注，并引起了一些开发人员和组织对其对项目治理、贡献动态和长期可持续性的影响的关注。目的：通过检查不同动机的开发人员的行为，分析Redis存储库中的贡献分布，评估许可证变更期间发生的变化，并评估云提供商是否在利用开源项目。方法：本研究基于协作行为数据对开发人员的动机进行分类。通过利用开发人员概要信息，贡献者被分为三种动机类型：公司驱动、社区驱动和通信驱动，从而可以分析不同贡献者群体的趋势。该研究主要依赖于社区健康分析开源软件（CHAOSS）社区的项目评估指标，以及协作指标，以检查Redis存储库中的变化。结果：自2017年以来，云提供商一直在为Redis开源项目做出贡献。在许可证变更期间，贡献者的参与和影响力显著下降，特别是在公司驱动和社区驱动的开发人员中。一些核心贡献者转移到新成立的Valkey项目。总结：在Redis Inc.更改许可证之后，社区经历了一定程度的分裂，主要的云提供商迁移到Valkey分支。云提供商已经认识到社区的重要性，并愿意将资源投入到他们参与的开源项目中，以确保更好的协作，并与上游云服务开发保持一致。

{"title":"Are cloud providers exploiting open-source? An exploratory study of Redis license change","authors":"Fanyu Han , Shengyu Zhao , Xiaoya Xia , Wei Wang","doi":"10.1016/j.infsof.2025.107951","DOIUrl":"10.1016/j.infsof.2025.107951","url":null,"abstract":"<div><h3>Context:</h3><div>On March 20, 2024, Redis Inc. changed the Redis project’s license from the permissive Berkeley Software Distribution (BSD) license to a dual licensing model under the Redis Source Available License (RSALv2) and the Server Side Public License (SSPLv1). The official rationale for this change was to restrict cloud service providers from offering Redis as a managed service without contributing back. The license transition drew widespread attention within the open source community and raised concerns from some developers and organizations about its implications for project governance, contribution dynamics, and long-term sustainability.</div></div><div><h3>Objective:</h3><div>Analyze the contribution distribution within the Redis repository by examining the behavior of developers with different motivations, evaluate the changes that occurred during the license change period, and assess whether cloud providers are exploiting the open-source project.</div></div><div><h3>Method:</h3><div>This study categorizes developers’ motivations based on collaboration behavior data. By leveraging developer profile information, contributors are categorized into three motivation types: company-driven, community-driven, and communication-driven, enabling an analysis of trends across different contributor groups. The study primarily relies on project evaluation metrics from the Community Health Analytics Open Source Software (CHAOSS) community, along with collaboration metrics, to examine changes within the Redis repository.</div></div><div><h3>Results:</h3><div>Since 2017, cloud providers have consistently contributed to the Redis open-source project. During the license change period, there was a notable decrease in contributor participation and influence, particularly among company-driven and community-driven developers. Several core contributors transitioned to the newly established Valkey project.</div></div><div><h3>Conclusion:</h3><div>Following the license change by Redis Inc., the community experienced a certain degree of fragmentation, with major cloud providers migrating to the Valkey fork. Cloud providers have recognized the importance of the community and are willing to invest resources into the open-source projects they participate in, ensuring better collaboration and alignment with upstream development for their cloud services.</div></div>","PeriodicalId":54983,"journal":{"name":"Information and Software Technology","volume":"190 ","pages":"Article 107951"},"PeriodicalIF":4.3,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145468018","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Software architecture for machine learning to aid sustainable digital transformation: A systematic mapping study 帮助可持续数字化转型的机器学习软件架构：系统映射研究

IF 4.3 2区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information and Software Technology

Pub Date : 2026-02-01 Epub Date: 2025-10-27 DOI: 10.1016/j.infsof.2025.107931

Lech Bialek , Rix Groenboom , Vasilios Andrikopoulos

Context:

Rapid developments and adoption of machine learning-based software solutions have enabled novel ways to tackle our societal problems. The ongoing digital transformation has led to the incorporation of these software solutions in just about every application domain. Software architecture for machine learning applications used during sustainable digital transformation can potentially aid the evolution of the underlying software system adding to its sustainability over time.

Objective:

Software architecture for machine learning applications in general is an open research area. When applying it to sustainable digital transformation it is not clear which of its considerations actually apply in this context. We therefore aim to understand how the topics of sustainable digital transformation, software architecture, and machine learning interact with each other.

Methods:

We perform a systematic mapping study to explore the scientific literature on the intersection of sustainable digital transformation, machine learning and software architecture.

Results:

We have found that the intersection of interest is small despite the amount of works on its individual aspects, and not all dimensions of sustainability are represented equally. We also found that application domains are diverse and include many important sectors and industry groups. At the same time, the perceived level of maturity of machine learning adoption by existing works seems to be quite low.

Conclusion:

Our findings show an opportunity for further software architecture research to aid sustainable digital transformation, especially by building on the emerging practice of machine learning operations.

背景：基于机器学习的软件解决方案的快速发展和采用，为解决我们的社会问题提供了新的方法。正在进行的数字化转型已经导致将这些软件解决方案整合到几乎每个应用程序领域。在可持续数字化转型期间使用的机器学习应用程序的软件架构可能有助于底层软件系统的发展，并随着时间的推移增加其可持续性。目的：机器学习应用的软件架构通常是一个开放的研究领域。在将其应用于可持续数字化转型时，尚不清楚哪些考虑因素实际适用于此背景。因此，我们的目标是了解可持续数字化转型、软件架构和机器学习等主题如何相互作用。方法：我们进行了系统的地图研究，以探索可持续数字化转型、机器学习和软件架构交叉的科学文献。结果：我们发现，尽管在其各个方面进行了大量的工作，但兴趣交集很小，并且并不是所有可持续性的维度都是平等的。我们还发现，应用程序领域是多种多样的，包括许多重要的部门和行业组。与此同时，现有作品对机器学习采用的成熟程度似乎相当低。结论：我们的研究结果表明，进一步的软件架构研究有机会帮助可持续的数字化转型，特别是通过建立在新兴的机器学习操作实践之上。

{"title":"Software architecture for machine learning to aid sustainable digital transformation: A systematic mapping study","authors":"Lech Bialek , Rix Groenboom , Vasilios Andrikopoulos","doi":"10.1016/j.infsof.2025.107931","DOIUrl":"10.1016/j.infsof.2025.107931","url":null,"abstract":"<div><h3>Context:</h3><div>Rapid developments and adoption of machine learning-based software solutions have enabled novel ways to tackle our societal problems. The ongoing digital transformation has led to the incorporation of these software solutions in just about every application domain. Software architecture for machine learning applications used during sustainable digital transformation can potentially aid the evolution of the underlying software system adding to its sustainability over time.</div></div><div><h3>Objective:</h3><div>Software architecture for machine learning applications in general is an open research area. When applying it to sustainable digital transformation it is not clear which of its considerations actually apply in this context. We therefore aim to understand how the topics of sustainable digital transformation, software architecture, and machine learning interact with each other.</div></div><div><h3>Methods:</h3><div>We perform a systematic mapping study to explore the scientific literature on the intersection of sustainable digital transformation, machine learning and software architecture.</div></div><div><h3>Results:</h3><div>We have found that the intersection of interest is small despite the amount of works on its individual aspects, and not all dimensions of sustainability are represented equally. We also found that application domains are diverse and include many important sectors and industry groups. At the same time, the perceived level of maturity of machine learning adoption by existing works seems to be quite low.</div></div><div><h3>Conclusion:</h3><div>Our findings show an opportunity for further software architecture research to aid sustainable digital transformation, especially by building on the emerging practice of machine learning operations.</div></div>","PeriodicalId":54983,"journal":{"name":"Information and Software Technology","volume":"190 ","pages":"Article 107931"},"PeriodicalIF":4.3,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145420382","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A systematic literature review on logging smell detection 对木材气味检测进行了系统的文献综述

IF 4.3 2区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information and Software Technology

Pub Date : 2026-02-01 Epub Date: 2025-11-05 DOI: 10.1016/j.infsof.2025.107961

Nora Madi, Manal Binkhonain

Context:

Logging is an important part of software development that helps developers monitor systems, understand behavior, and fix problems. But when logging is done poorly, it can introduce logging smells, which are defects that reduce the usefulness of logs or even make them problematic.

Objective:

This study looks at how logging smells are currently detected. The goal is to better understand the existing research on automatic detection techniques, datasets, and evaluation methods.

Method:

We conducted a systematic literature review (SLR) of 21 studies focused on detecting logging smells. In this review, we define key logging-related terms, identify and map the types of smells to an existing taxonomy, and examine the detection techniques, datasets, and evaluation strategies used across the studies.

Results:

We found that the research is still scattered and inconsistent. For example, there is no common benchmark or standardized approach for evaluating results, making it difficult to compare studies. In addition, we observe inconsistencies in the way log smells are addressed, as studies differ in the types and number of smells they target.

Conclusion:

There is still room for improvement in how logging smells are studied and detected. We point out several challenges and suggest future directions, such as developing better tools, using large language models (LLMs), and building more standardized datasets for evaluation.

上下文：日志记录是软件开发的一个重要部分，它可以帮助开发人员监视系统、理解行为和修复问题。但是，如果日志记录做得不好，就会引入日志气味，这些缺陷会降低日志的有用性，甚至使日志出现问题。目的：本研究着眼于当前如何检测伐木气味。目的是更好地理解现有的自动检测技术、数据集和评估方法的研究。方法：对21篇有关木材气味检测的研究进行了系统的文献综述。在这篇综述中，我们定义了与日志记录相关的关键术语，识别并将气味类型映射到现有的分类法中，并检查了研究中使用的检测技术、数据集和评估策略。结果：我们发现研究仍然是分散和不一致的。例如，没有共同的基准或标准化的方法来评估结果，这使得比较研究变得困难。此外，我们观察到日志气味处理方式的不一致性，因为研究的目标气味类型和数量不同。结论：木材气味的研究和检测仍有改进的空间。我们指出了几个挑战，并提出了未来的方向，例如开发更好的工具，使用大型语言模型（llm），以及构建更标准化的数据集进行评估。

{"title":"A systematic literature review on logging smell detection","authors":"Nora Madi, Manal Binkhonain","doi":"10.1016/j.infsof.2025.107961","DOIUrl":"10.1016/j.infsof.2025.107961","url":null,"abstract":"<div><h3>Context:</h3><div>Logging is an important part of software development that helps developers monitor systems, understand behavior, and fix problems. But when logging is done poorly, it can introduce logging smells, which are defects that reduce the usefulness of logs or even make them problematic.</div></div><div><h3>Objective:</h3><div>This study looks at how logging smells are currently detected. The goal is to better understand the existing research on automatic detection techniques, datasets, and evaluation methods.</div></div><div><h3>Method:</h3><div>We conducted a systematic literature review (SLR) of 21 studies focused on detecting logging smells. In this review, we define key logging-related terms, identify and map the types of smells to an existing taxonomy, and examine the detection techniques, datasets, and evaluation strategies used across the studies.</div></div><div><h3>Results:</h3><div>We found that the research is still scattered and inconsistent. For example, there is no common benchmark or standardized approach for evaluating results, making it difficult to compare studies. In addition, we observe inconsistencies in the way log smells are addressed, as studies differ in the types and number of smells they target.</div></div><div><h3>Conclusion:</h3><div>There is still room for improvement in how logging smells are studied and detected. We point out several challenges and suggest future directions, such as developing better tools, using large language models (LLMs), and building more standardized datasets for evaluation.</div></div>","PeriodicalId":54983,"journal":{"name":"Information and Software Technology","volume":"190 ","pages":"Article 107961"},"PeriodicalIF":4.3,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145520801","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The Software Diversity Card: A framework for reporting diversity in software projects 软件多样性卡：一个报告软件项目多样性的框架

IF 4.3 2区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information and Software Technology

Pub Date : 2026-02-01 Epub Date: 2025-11-01 DOI: 10.1016/j.infsof.2025.107950

Joan Giner-Miguelez , Sergio Morales , Sergio Cobos , Javier Luis Cánovas Izquierdo , Robert Clarisó , Jordi Cabot

Context:

Interest in diversity in software development has significantly increased in recent years. Reporting on diversity in software projects can enhance user trust and assist regulators in evaluating adoption. Recent AI directives include clauses that mandate diversity information during development, highlighting the growing interest of public regulators. However, current documentation often neglects diversity in favor of technical features, partly due to a lack of tools for its description and annotation.

Objectives:

This work introduces the Software Diversity Card, a structured approach to document and share diversity-related aspects within software projects. It aims to profile the various teams involved in software development and governance, including user groups in testing and software adaptations for diverse social groups.

Methods:

We performed a literature review on diversity and inclusion in software development and an analysis of 1000 top-starred Open Source Software (OSS) repositories in GitHub to identify diversity-related information. Moreover, we present a diversity modeling language, a toolkit for generating the cards using that language, and a study of its application in two real-world software projects.

Results:

Despite the growing awareness of diversity in the research community, our analysis found a notable lack of diversity reporting in OSS projects. Applying the card to real-world examples highlighted challenges like balancing anonymity with transparency, managing sensitive data, and ensuring authenticity.

Conclusion:

We believe that our proposal can enhance diversity practices in software development, support public administrations in software assessment, and help businesses promote diversity as a key asset.

背景：近年来，对软件开发多样性的兴趣显著增加。报告软件项目中的多样性可以增强用户信任，并帮助管理者评估采用情况。最近的人工智能指令包括要求在开发过程中提供多样化信息的条款，突显了公共监管机构日益增长的兴趣。然而，当前的文档常常忽略了技术特性的多样性，部分原因是缺乏用于描述和注释的工具。目标：本工作介绍了软件多样性卡，这是一种在软件项目中记录和共享多样性相关方面的结构化方法。它旨在描述参与软件开发和治理的各种团队，包括测试中的用户组和针对不同社会群体的软件调整。方法：我们对软件开发中的多样性和包容性进行文献综述，并对GitHub上1000个星级开源软件（OSS）库进行分析，以识别与多样性相关的信息。此外，我们提出了一种多样性建模语言，一个使用该语言生成卡片的工具包，并研究了它在两个实际软件项目中的应用。结果：尽管在研究社区中对多样性的认识越来越高，我们的分析发现在OSS项目中明显缺乏多样性报告。将该卡应用于现实世界的例子凸显了平衡匿名性与透明度、管理敏感数据和确保真实性等挑战。结论：我们相信我们的建议可以增强软件开发中的多样性实践，支持软件评估中的公共管理，并帮助企业将多样性作为关键资产来促进。

{"title":"The Software Diversity Card: A framework for reporting diversity in software projects","authors":"Joan Giner-Miguelez , Sergio Morales , Sergio Cobos , Javier Luis Cánovas Izquierdo , Robert Clarisó , Jordi Cabot","doi":"10.1016/j.infsof.2025.107950","DOIUrl":"10.1016/j.infsof.2025.107950","url":null,"abstract":"<div><h3>Context:</h3><div>Interest in diversity in software development has significantly increased in recent years. Reporting on diversity in software projects can enhance user trust and assist regulators in evaluating adoption. Recent AI directives include clauses that mandate diversity information during development, highlighting the growing interest of public regulators. However, current documentation often neglects diversity in favor of technical features, partly due to a lack of tools for its description and annotation.</div></div><div><h3>Objectives:</h3><div>This work introduces the Software Diversity Card, a structured approach to document and share diversity-related aspects within software projects. It aims to profile the various teams involved in software development and governance, including user groups in testing and software adaptations for diverse social groups.</div></div><div><h3>Methods:</h3><div>We performed a literature review on diversity and inclusion in software development and an analysis of 1000 top-starred Open Source Software (OSS) repositories in <span>GitHub</span> to identify diversity-related information. Moreover, we present a diversity modeling language, a toolkit for generating the cards using that language, and a study of its application in two real-world software projects.</div></div><div><h3>Results:</h3><div>Despite the growing awareness of diversity in the research community, our analysis found a notable lack of diversity reporting in OSS projects. Applying the card to real-world examples highlighted challenges like balancing anonymity with transparency, managing sensitive data, and ensuring authenticity.</div></div><div><h3>Conclusion:</h3><div>We believe that our proposal can enhance diversity practices in software development, support public administrations in software assessment, and help businesses promote diversity as a key asset.</div></div>","PeriodicalId":54983,"journal":{"name":"Information and Software Technology","volume":"190 ","pages":"Article 107950"},"PeriodicalIF":4.3,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145520804","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SABLM-VD: Vulnerability detection with a semantic-aware binary language model SABLM-VD：使用语义感知的二进制语言模型进行漏洞检测

IF 4.3 2区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information and Software Technology

Pub Date : 2026-02-01 Epub Date: 2025-11-12 DOI: 10.1016/j.infsof.2025.107959

Qinghao Li, Tieming Liu, Wei Liu, Yonghe Tang, Chunling Liu, Weiyu Dong

Context:

Static detection of binary code vulnerabilities based on deep learning is an important research field in computer security.

Objective:

In existing methods, those mainly based on code similarity often focus on overall similarity while neglecting crucial and subtle semantic differences in the code, leading to potential false positives. Meanwhile, methods mainly based on vulnerability patterns still face challenges in learning semantic features.

Method:

To address the above problems, we conduct research on x86-64 and ARM architecture binaries and propose a hybrid-granularity assembly language tokenization method. Then, we propose a BERT-based semantic-aware binary language model, SABLM. It is a solution that simultaneously embeds the data transfer semantics, arithmetic logic semantics, and control flow of assembly instructions into a BERT-based language model, effectively perceiving the semantics of assembly code. Based on the semantic feature representation of assembly code by SABLM, we incorporate pseudocode features and construct a cross-architecture vulnerability detection framework, SABLM-VD.

Results:

We evaluate SABLM-VD on the NVD dataset and the SARD dataset. The results show that SABLM-VD outperforms the state-of-the-art baseline methods in terms of F1-score, precision, recall, and accuracy, with SABLM-VD achieving F1-scores of 79.33% on NVD (Mixed), 100.00% on SARD (x86-64), and 100.00% on SARD (ARM). Ablation studies and component analysis demonstrate the effectiveness of each component of SABLM-VD. Visualizations and real-world applications further confirm the advantages of SABLM-VD.

Conclusion:

Our research indicates that SABLM-VD, which is based on the semantic-aware binary language model SABLM and pseudocode features, can effectively detect binary code vulnerabilities, warranting further research in this direction.

背景：基于深度学习的二进制代码漏洞静态检测是计算机安全领域的一个重要研究领域。目的：在现有的方法中，主要基于代码相似度的方法往往只关注整体的相似度，而忽略了代码中关键和微妙的语义差异，从而导致潜在的误报。同时，主要基于漏洞模式的方法在语义特征学习方面仍然面临挑战。方法：针对以上问题，对x86-64和ARM架构的二进制文件进行研究，提出一种混合粒度的汇编语言标记方法。然后，我们提出了一个基于bert的语义感知二进制语言模型SABLM。该解决方案将汇编指令的数据传输语义、算术逻辑语义和控制流同时嵌入到基于bert的语言模型中，从而有效地感知汇编代码的语义。基于SABLM对汇编代码的语义特征表示，结合伪代码特征，构建了一个跨体系结构的漏洞检测框架SABLM- vd。结果：我们在NVD数据集和SARD数据集上对SABLM-VD进行了评估。结果表明，SABLM-VD在f1评分、精密度、召回率和准确率方面都优于目前最先进的基线方法，其中SABLM-VD在NVD （Mixed）上的f1得分为79.33%，在SARD （x86-64）上的f1得分为100.00%，在SARD （ARM）上的f1得分为100.00%。消融研究和成分分析证明了SABLM-VD各成分的有效性。可视化和实际应用进一步证实了SABLM-VD的优势。结论：我们的研究表明，基于语义感知二进制语言模型SABLM和伪代码特征的SABLM- vd可以有效地检测二进制代码漏洞，值得进一步研究。

{"title":"SABLM-VD: Vulnerability detection with a semantic-aware binary language model","authors":"Qinghao Li, Tieming Liu, Wei Liu, Yonghe Tang, Chunling Liu, Weiyu Dong","doi":"10.1016/j.infsof.2025.107959","DOIUrl":"10.1016/j.infsof.2025.107959","url":null,"abstract":"<div><h3>Context:</h3><div>Static detection of binary code vulnerabilities based on deep learning is an important research field in computer security.</div></div><div><h3>Objective:</h3><div>In existing methods, those mainly based on code similarity often focus on overall similarity while neglecting crucial and subtle semantic differences in the code, leading to potential false positives. Meanwhile, methods mainly based on vulnerability patterns still face challenges in learning semantic features.</div></div><div><h3>Method:</h3><div>To address the above problems, we conduct research on x86-64 and ARM architecture binaries and propose a hybrid-granularity assembly language tokenization method. Then, we propose a BERT-based semantic-aware binary language model, SABLM. It is a solution that simultaneously embeds the data transfer semantics, arithmetic logic semantics, and control flow of assembly instructions into a BERT-based language model, effectively perceiving the semantics of assembly code. Based on the semantic feature representation of assembly code by SABLM, we incorporate pseudocode features and construct a cross-architecture vulnerability detection framework, SABLM-VD.</div></div><div><h3>Results:</h3><div>We evaluate SABLM-VD on the NVD dataset and the SARD dataset. The results show that SABLM-VD outperforms the state-of-the-art baseline methods in terms of F1-score, precision, recall, and accuracy, with SABLM-VD achieving F1-scores of 79.33% on NVD (Mixed), 100.00% on SARD (x86-64), and 100.00% on SARD (ARM). Ablation studies and component analysis demonstrate the effectiveness of each component of SABLM-VD. Visualizations and real-world applications further confirm the advantages of SABLM-VD.</div></div><div><h3>Conclusion:</h3><div>Our research indicates that SABLM-VD, which is based on the semantic-aware binary language model SABLM and pseudocode features, can effectively detect binary code vulnerabilities, warranting further research in this direction.</div></div>","PeriodicalId":54983,"journal":{"name":"Information and Software Technology","volume":"190 ","pages":"Article 107959"},"PeriodicalIF":4.3,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145520805","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

DPDGPT: Using Multimodal Large Language Models for automated detection of dark patterns DPDGPT：使用多模态大语言模型自动检测暗模式

IF 4.3 2区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information and Software Technology

Pub Date : 2026-02-01 Epub Date: 2025-11-04 DOI: 10.1016/j.infsof.2025.107936

Fengwei Lin , Liming Nie , Lei Xue , Xiaoxi Zhang , Kelei Zhang

Context:

Dark patterns are manipulative design strategies in user interfaces (UIs) that deceive or pressure users into taking unintended actions. These deceptive practices are increasingly prevalent in digital environments, affecting user trust and leading to unethical outcomes. Traditional dark pattern detection methods face challenges in scalability, adaptability, and explainability, necessitating more sophisticated and automated solutions.

Objective:

This study proposes DPDGPT, a novel framework that leverages the advanced reasoning capabilities of Multimodal Large Language Models for the automated screening of dark patterns in user interfaces. The goal is to improve the scale and reliability of automated screening, adaptability to diverse UI designs, and the interpretability of detection results.

Methods:

DPDGPT employs a multi-stage pipeline: Visual and Text Element Extraction, Initial Inference, Verification, and Advanced Inference. It extracts visual and textual features from the UIs, infers potential dark patterns using predefined knowledge, verifies these inferences against known features, and performs a detailed analysis for complex cases.

Results:

DPDGPT was evaluated using the ContextRico-DP dataset, which consists of 1524 mobile UIs containing 1871 instances across 13 dark pattern types, and 85 web UIs with 144 instances across 8 dark pattern types. DPDGPT achieved an average precision of 0.86, an average recall of 0.91, and an average F1 score of 0.88, successfully identifying 19 dark pattern types. It notably outperformed existing methods by achieving greater recall and a better balance between precision and recall.

Conclusion:

DPDGPT offers an efficient, adaptable solution for detecting diverse dark patterns in mobile and web UIs.The integrated multimodal approach offers strong potential for practical application in promoting ethical and user-friendly UI designs across digital platforms. Future work will focus on expanding the coverage of the platform and improving the detection of subtle patterns.

背景：暗模式是用户界面（ui）中的操纵性设计策略，它欺骗或迫使用户采取意想不到的行动。这些欺骗行为在数字环境中越来越普遍，影响用户信任并导致不道德的结果。传统的暗模式检测方法在可扩展性、适应性和可解释性方面面临挑战，需要更复杂和自动化的解决方案。目的：本研究提出了DPDGPT，这是一个利用多模态大语言模型的高级推理能力来自动筛选用户界面中的暗模式的新框架。目标是提高自动化筛选的规模和可靠性，对不同UI设计的适应性，以及检测结果的可解释性。方法：DPDGPT采用多阶段流程：视觉和文本元素提取、初始推理、验证和高级推理。它从ui中提取视觉和文本特征，使用预定义的知识推断潜在的暗模式，根据已知特征验证这些推断，并对复杂情况执行详细分析。结果：DPDGPT使用contextreco - dp数据集进行评估，该数据集由1524个移动ui组成，包含13种暗模式类型的1871个实例，以及85个web ui，包含8种暗模式类型的144个实例。DPDGPT的平均准确率为0.86，平均召回率为0.91，平均F1得分为0.88，成功识别了19种暗纹类型。它通过实现更高的召回率和更好的准确率和召回率之间的平衡，明显优于现有的方法。结论：DPDGPT是一种高效、适应性强的方法，可用于检测移动和网络用户界面中的各种暗模式。综合多模式方法在促进跨数字平台的道德和用户友好的UI设计方面具有强大的实际应用潜力。未来的工作将集中在扩大平台的覆盖范围和提高对细微模式的检测。

{"title":"DPDGPT: Using Multimodal Large Language Models for automated detection of dark patterns","authors":"Fengwei Lin , Liming Nie , Lei Xue , Xiaoxi Zhang , Kelei Zhang","doi":"10.1016/j.infsof.2025.107936","DOIUrl":"10.1016/j.infsof.2025.107936","url":null,"abstract":"<div><h3>Context:</h3><div>Dark patterns are manipulative design strategies in user interfaces (UIs) that deceive or pressure users into taking unintended actions. These deceptive practices are increasingly prevalent in digital environments, affecting user trust and leading to unethical outcomes. Traditional dark pattern detection methods face challenges in scalability, adaptability, and explainability, necessitating more sophisticated and automated solutions.</div></div><div><h3>Objective:</h3><div>This study proposes DPDGPT, a novel framework that leverages the advanced reasoning capabilities of Multimodal Large Language Models for the automated screening of dark patterns in user interfaces. The goal is to improve the scale and reliability of automated screening, adaptability to diverse UI designs, and the interpretability of detection results.</div></div><div><h3>Methods:</h3><div>DPDGPT employs a multi-stage pipeline: Visual and Text Element Extraction, Initial Inference, Verification, and Advanced Inference. It extracts visual and textual features from the UIs, infers potential dark patterns using predefined knowledge, verifies these inferences against known features, and performs a detailed analysis for complex cases.</div></div><div><h3>Results:</h3><div>DPDGPT was evaluated using the ContextRico-DP dataset, which consists of 1524 mobile UIs containing 1871 instances across 13 dark pattern types, and 85 web UIs with 144 instances across 8 dark pattern types. DPDGPT achieved an average precision of 0.86, an average recall of 0.91, and an average F1 score of 0.88, successfully identifying 19 dark pattern types. It notably outperformed existing methods by achieving greater recall and a better balance between precision and recall.</div></div><div><h3>Conclusion:</h3><div>DPDGPT offers an efficient, adaptable solution for detecting diverse dark patterns in mobile and web UIs.The integrated multimodal approach offers strong potential for practical application in promoting ethical and user-friendly UI designs across digital platforms. Future work will focus on expanding the coverage of the platform and improving the detection of subtle patterns.</div></div>","PeriodicalId":54983,"journal":{"name":"Information and Software Technology","volume":"190 ","pages":"Article 107936"},"PeriodicalIF":4.3,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145520799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

UF-CDDFM: A unified framework for code defect detection using multi-modal inputs and few-shot learning UF-CDDFM：使用多模态输入和少量学习的代码缺陷检测的统一框架

IF 4.3 2区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information and Software Technology

Pub Date : 2026-02-01 Epub Date: 2025-10-29 DOI: 10.1016/j.infsof.2025.107942

Xianglu Zhou , Tianxiang Cui , Xiaoyan Zhu , Jiayin Wang , Xin Lai

Context:

The detection of code defects is foundational to modern software development and maintenance, playing a critical role in ensuring software quality and security. However, as software systems grow in scale and complexity, the limitations of traditional static analysis and conventional machine learning techniques have become increasingly evident. These methods rely heavily on intricate, manual feature engineering and fail to capture dynamic runtime behavior, resulting in suboptimal accuracy and elevated error rates.

Objective:

To address these deficiencies, we propose UF-CDDFM, a unified framework for code defect detection that integrates multi-modal inputs, active learning, and state-of-the-art few-shot learning techniques. We aim to improve detection performance, reduce feature selection complexity and sample bias through active learning, and maintain practical efficiency in real-world development contexts.

Methods:

UF-CDDFM employs parallel encoding of source code, code annotations, and abstract syntax trees (ASTs) using large language models (LLMs) alongside multilayer perceptrons (MLPs) to derive robust, high-fidelity representations of code. To streamline feature selection and mitigate sample bias, an active learning component is introduced for automated identification of high-quality features. Addressing the pervasive challenge of data scarcity, we incorporate two complementary few-shot learning strategies-MAML for small-scale datasets and LEO for larger-scale settings to enhance overall generalization capability.

Results:

Empirical evaluations demonstrate that UF-CDDFM consistently outperforms existing methods, establishing new state-of-the-art detection rates: 72.04% for defect detection and 95.23% for clone detection. Crucially, these gains are achieved within resource-constrained computational environments, which highlights the practicality of the method.

Conclusion:

By fusing multi-modal code representations, active learning, and adaptive few-shot learning techniques, UF-CDDFM delivers significant improvements in detection accuracy and computational efficiency. This work offers a new paradigm for robust, scalable, and practical code defect and clone detection in modern software engineering.

背景：代码缺陷的检测是现代软件开发和维护的基础，在确保软件质量和安全性方面起着至关重要的作用。然而，随着软件系统规模和复杂性的增长，传统静态分析和传统机器学习技术的局限性变得越来越明显。这些方法严重依赖于复杂的手动特征工程，无法捕获动态运行时行为，从而导致次优精度和错误率升高。目的：为了解决这些缺陷，我们提出了UF-CDDFM，这是一个统一的代码缺陷检测框架，它集成了多模态输入、主动学习和最先进的少量学习技术。我们的目标是通过主动学习提高检测性能，降低特征选择的复杂性和样本偏差，并在现实世界的开发环境中保持实际的效率。方法：UF-CDDFM采用源代码、代码注释和抽象语法树（ast）的并行编码，使用大型语言模型（llm）和多层感知器（mlp）来获得代码的鲁棒性、高保真度表示。为了简化特征选择和减轻样本偏差，引入了主动学习组件来自动识别高质量的特征。为了解决普遍存在的数据稀缺性挑战，我们结合了两种互补的少量学习策略——用于小规模数据集的maml和用于大规模设置的LEO，以增强整体泛化能力。结果：经验评估表明，UF-CDDFM始终优于现有方法，建立了新的最先进的检测率：缺陷检测率为72.04%，克隆检测率为95.23%。至关重要的是，这些收益是在资源受限的计算环境中实现的，这突出了该方法的实用性。结论：通过融合多模态代码表示、主动学习和自适应少镜头学习技术，UF-CDDFM在检测精度和计算效率方面有显著提高。这项工作为现代软件工程中健壮的、可伸缩的、实用的代码缺陷和克隆检测提供了一个新的范例。

{"title":"UF-CDDFM: A unified framework for code defect detection using multi-modal inputs and few-shot learning","authors":"Xianglu Zhou , Tianxiang Cui , Xiaoyan Zhu , Jiayin Wang , Xin Lai","doi":"10.1016/j.infsof.2025.107942","DOIUrl":"10.1016/j.infsof.2025.107942","url":null,"abstract":"<div><h3>Context:</h3><div>The detection of code defects is foundational to modern software development and maintenance, playing a critical role in ensuring software quality and security. However, as software systems grow in scale and complexity, the limitations of traditional static analysis and conventional machine learning techniques have become increasingly evident. These methods rely heavily on intricate, manual feature engineering and fail to capture dynamic runtime behavior, resulting in suboptimal accuracy and elevated error rates.</div></div><div><h3>Objective:</h3><div>To address these deficiencies, we propose UF-CDDFM, a unified framework for code defect detection that integrates multi-modal inputs, active learning, and state-of-the-art few-shot learning techniques. We aim to improve detection performance, reduce feature selection complexity and sample bias through active learning, and maintain practical efficiency in real-world development contexts.</div></div><div><h3>Methods:</h3><div>UF-CDDFM employs parallel encoding of source code, code annotations, and abstract syntax trees (ASTs) using large language models (LLMs) alongside multilayer perceptrons (MLPs) to derive robust, high-fidelity representations of code. To streamline feature selection and mitigate sample bias, an active learning component is introduced for automated identification of high-quality features. Addressing the pervasive challenge of data scarcity, we incorporate two complementary few-shot learning strategies-MAML for small-scale datasets and LEO for larger-scale settings to enhance overall generalization capability.</div></div><div><h3>Results:</h3><div>Empirical evaluations demonstrate that UF-CDDFM consistently outperforms existing methods, establishing new state-of-the-art detection rates: 72.04% for defect detection and 95.23% for clone detection. Crucially, these gains are achieved within resource-constrained computational environments, which highlights the practicality of the method.</div></div><div><h3>Conclusion:</h3><div>By fusing multi-modal code representations, active learning, and adaptive few-shot learning techniques, UF-CDDFM delivers significant improvements in detection accuracy and computational efficiency. This work offers a new paradigm for robust, scalable, and practical code defect and clone detection in modern software engineering.</div></div>","PeriodicalId":54983,"journal":{"name":"Information and Software Technology","volume":"190 ","pages":"Article 107942"},"PeriodicalIF":4.3,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145420380","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

DSL-Xpert 2.0: Enhancing LLM-driven code generation for domain-specific languages DSL-Xpert 2.0：增强针对领域特定语言的法学硕士驱动的代码生成

IF 4.3 2区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information and Software Technology

Pub Date : 2026-02-01 Epub Date: 2025-10-31 DOI: 10.1016/j.infsof.2025.107954

Victor Lamas , Daniel Garcia-Gonzalez , Luca Sala , Miguel R. Luaces

Context:

Domain-specific languages (DSLs) are essential for modeling specialized concepts, offering greater fluency and efficiency than general-purpose languages. However, their adoption is often hindered by steep learning curves, limited tools, and complex implementations. While large language models (LLMs) can generate DSL code from natural language, their performance is limited in niche areas due to a lack of training on specific DSL definitions.

Objectives:

This paper introduces DSL-Xpert 2.0, a tool that addresses these challenges by using LLMs to generate DSL code effortlessly.

Methods:

Integrating grammar prompting and few-shot learning ensures the effective handling of proprietary DSLs. In addition, advanced features such as automatic grammar validation, input/output correction, and integration with platforms like OpenAI, HuggingFace, and WebLLM provide robust, reliable results while simplifying workflows for novices and experts. To further demonstrate the tool’s practical value, this paper provides a running example illustrating its workflow and a complementary user survey conducted across multiple DSLs of varying complexity, following the Technology Acceptance Model (TAM), to evaluate its impact on easing the DSL learning curve.

Results:

With a user-friendly and flexible design, DSL-Xpert 2.0 supports a wide range of DSL designs with minimal configuration. Its intuitive interface allows developers to focus on innovative problem-solving rather than technical complexities. Findings from the user survey confirm that DSL-Xpert 2.0 effectively reduces the learning effort required to work with DSLs and is perceived as both useful and easy to use. Additionally, this paper provides a detailed performance analysis across various LLMs, showcasing the adaptability and effectiveness of the tool.

Conclusion:

By simplifying DSL development and lowering entry barriers, DSL-Xpert 2.0 accelerates adoption and innovation, positioning itself as a valuable resource for domain-specific projects.

上下文：领域特定语言（dsl）对于专门概念的建模是必不可少的，它比通用语言提供了更高的流畅性和效率。然而，它们的采用常常受到陡峭的学习曲线、有限的工具和复杂的实现的阻碍。虽然大型语言模型（llm）可以从自然语言生成DSL代码，但由于缺乏对特定DSL定义的培训，它们的性能在特定领域受到限制。目标：本文介绍了DSL- xpert 2.0，这是一个通过使用llm轻松生成DSL代码来解决这些挑战的工具。方法：将语法提示和少次学习相结合，保证了专有领域特定语言的有效处理。此外，自动语法验证、输入/输出校正以及与OpenAI、HuggingFace和WebLLM等平台的集成等高级功能提供了强大、可靠的结果，同时简化了新手和专家的工作流程。为了进一步展示该工具的实用价值，本文提供了一个运行的示例来说明其工作流程，并根据技术接受模型（TAM）在多个不同复杂性的DSL中进行了补充用户调查，以评估其对缓解DSL学习曲线的影响。结果：DSL- xpert 2.0具有用户友好和灵活的设计，以最小的配置支持广泛的DSL设计。其直观的界面允许开发人员专注于创新的问题解决，而不是技术复杂性。用户调查的结果证实，DSL-Xpert 2.0有效地减少了使用dsl所需的学习工作量，并且被认为既有用又易于使用。此外，本文还提供了各种llm的详细性能分析，展示了该工具的适应性和有效性。结论：通过简化DSL开发和降低进入门槛，DSL- xpert 2.0加速了采用和创新，将自己定位为领域特定项目的宝贵资源。

{"title":"DSL-Xpert 2.0: Enhancing LLM-driven code generation for domain-specific languages","authors":"Victor Lamas , Daniel Garcia-Gonzalez , Luca Sala , Miguel R. Luaces","doi":"10.1016/j.infsof.2025.107954","DOIUrl":"10.1016/j.infsof.2025.107954","url":null,"abstract":"<div><h3>Context:</h3><div>Domain-specific languages (DSLs) are essential for modeling specialized concepts, offering greater fluency and efficiency than general-purpose languages. However, their adoption is often hindered by steep learning curves, limited tools, and complex implementations. While large language models (LLMs) can generate DSL code from natural language, their performance is limited in niche areas due to a lack of training on specific DSL definitions.</div></div><div><h3>Objectives:</h3><div>This paper introduces DSL-Xpert 2.0, a tool that addresses these challenges by using LLMs to generate DSL code effortlessly.</div></div><div><h3>Methods:</h3><div>Integrating grammar prompting and few-shot learning ensures the effective handling of proprietary DSLs. In addition, advanced features such as automatic grammar validation, input/output correction, and integration with platforms like OpenAI, HuggingFace, and WebLLM provide robust, reliable results while simplifying workflows for novices and experts. To further demonstrate the tool’s practical value, this paper provides a running example illustrating its workflow and a complementary user survey conducted across multiple DSLs of varying complexity, following the Technology Acceptance Model (TAM), to evaluate its impact on easing the DSL learning curve.</div></div><div><h3>Results:</h3><div>With a user-friendly and flexible design, DSL-Xpert 2.0 supports a wide range of DSL designs with minimal configuration. Its intuitive interface allows developers to focus on innovative problem-solving rather than technical complexities. Findings from the user survey confirm that DSL-Xpert 2.0 effectively reduces the learning effort required to work with DSLs and is perceived as both useful and easy to use. Additionally, this paper provides a detailed performance analysis across various LLMs, showcasing the adaptability and effectiveness of the tool.</div></div><div><h3>Conclusion:</h3><div>By simplifying DSL development and lowering entry barriers, DSL-Xpert 2.0 accelerates adoption and innovation, positioning itself as a valuable resource for domain-specific projects.</div></div>","PeriodicalId":54983,"journal":{"name":"Information and Software Technology","volume":"190 ","pages":"Article 107954"},"PeriodicalIF":4.3,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145468023","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

RLSPL: A software product line for streamlining reinforcement learning project development RLSPL：简化强化学习项目开发的软件产品线

IF 4.3 2区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information and Software Technology

Pub Date : 2026-02-01 Epub Date: 2025-10-22 DOI: 10.1016/j.infsof.2025.107916

Syrine Wardi , Rania Mzid , Tewfik Ziadi

Context:

Reinforcement Learning (RL) is a powerful machine learning paradigm where agents learn optimal behaviors by interacting with dynamic environments. The development of RL systems is intricate and resource-intensive due to significant variability in algorithm choices, hyperparameters, and implementation intricacies.

Objectives:

This research aims to evaluate the potential of Software Product Line Engineering (SPLE) to simplify and streamline the RL development process through systematic reuse and customization.

Methods:

We developed a Software Product Line (SPL) named RLSPL, specifically tailored for RL applications. RLSPL employs a feature-oriented architecture that encapsulates essential RL components, allowing developers to systematically generate customized software variants. To evaluate RLSPL’s effectiveness, we applied it to various RL scenarios, assessing the ease of customization, code maintainability, and efficiency of variant generation.

Results:

Experimental evaluations demonstrated that RLSPL significantly simplifies the customization and reuse of RL solutions. The structured reuse of core RL assets improved maintainability and allowed rapid generation of software variants tailored to distinct scenarios.

Conclusions:

Our findings highlight the advantages of applying SPLE techniques to RL. RLSPL effectively addresses variability challenges inherent in RL development by providing structured reuse and automated customization. This structured approach not only enhances productivity but also supports consistent implementation quality. Across the evaluated variants, derivation reused over 95% of the code and preserved high coverage and maintainability The study confirms the feasibility and benefits of integrating SPLE principles into RL practices, highlighting RLSPL’s potential to standardize and simplify complex RL development processes.

背景：强化学习（RL）是一种强大的机器学习范式，智能体通过与动态环境交互来学习最佳行为。由于算法选择、超参数和实现复杂性的显著可变性，强化学习系统的开发是复杂和资源密集型的。目的：本研究旨在评估软件产品线工程（SPLE）通过系统重用和定制简化和流线化RL开发过程的潜力。方法：我们开发了一个名为RLSPL的软件产品线（SPL），专门为RL应用量身定制。RLSPL采用面向特性的架构，封装了必要的RL组件，允许开发人员系统地生成定制的软件变体。为了评估RLSPL的有效性，我们将其应用于各种RL场景，评估自定义的容易程度、代码的可维护性和变体生成的效率。结果：实验评估表明，RLSPL显著简化了RL解决方案的定制和重用。核心RL资产的结构化重用提高了可维护性，并允许快速生成针对不同场景的软件变体。结论：我们的研究结果突出了将SPLE技术应用于RL的优势。RLSPL通过提供结构化重用和自动化定制，有效地解决了RL开发中固有的可变性挑战。这种结构化的方法不仅提高了生产力，而且支持一致的实现质量。在评估的变体中，派生重用了95%以上的代码，并保持了高覆盖率和可维持性。该研究证实了将SPLE原则集成到RL实践中的可行性和好处，突出了RLSPL标准化和简化复杂RL开发过程的潜力。

{"title":"RLSPL: A software product line for streamlining reinforcement learning project development","authors":"Syrine Wardi , Rania Mzid , Tewfik Ziadi","doi":"10.1016/j.infsof.2025.107916","DOIUrl":"10.1016/j.infsof.2025.107916","url":null,"abstract":"<div><h3>Context:</h3><div>Reinforcement Learning (RL) is a powerful machine learning paradigm where agents learn optimal behaviors by interacting with dynamic environments. The development of RL systems is intricate and resource-intensive due to significant variability in algorithm choices, hyperparameters, and implementation intricacies.</div></div><div><h3>Objectives:</h3><div>This research aims to evaluate the potential of Software Product Line Engineering (SPLE) to simplify and streamline the RL development process through systematic reuse and customization.</div></div><div><h3>Methods:</h3><div>We developed a Software Product Line (SPL) named RLSPL, specifically tailored for RL applications. RLSPL employs a feature-oriented architecture that encapsulates essential RL components, allowing developers to systematically generate customized software variants. To evaluate RLSPL’s effectiveness, we applied it to various RL scenarios, assessing the ease of customization, code maintainability, and efficiency of variant generation.</div></div><div><h3>Results:</h3><div>Experimental evaluations demonstrated that RLSPL significantly simplifies the customization and reuse of RL solutions. The structured reuse of core RL assets improved maintainability and allowed rapid generation of software variants tailored to distinct scenarios.</div></div><div><h3>Conclusions:</h3><div>Our findings highlight the advantages of applying SPLE techniques to RL. RLSPL effectively addresses variability challenges inherent in RL development by providing structured reuse and automated customization. This structured approach not only enhances productivity but also supports consistent implementation quality. Across the evaluated variants, derivation reused over 95% of the code and preserved high coverage and maintainability The study confirms the feasibility and benefits of integrating SPLE principles into RL practices, highlighting RLSPL’s potential to standardize and simplify complex RL development processes.</div></div>","PeriodicalId":54983,"journal":{"name":"Information and Software Technology","volume":"190 ","pages":"Article 107916"},"PeriodicalIF":4.3,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145371146","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Accessibility evaluation of interaction modalities and cognitive process of self-service technologies’ user interface in Japan 日本自助服务技术用户界面交互方式与认知过程的可达性评价

IF 4.3 2区计算机科学 Q2 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information and Software Technology

Pub Date : 2026-02-01 Epub Date: 2025-11-11 DOI: 10.1016/j.infsof.2025.107970

Juho-Pekka Mäkipää , Junichi Iijima

As more and more services are becoming available only in digital form, self-service technologies (SSTs) need to be accessible to ensure all citizens have equal opportunities to participate in society. However, SSTs’ accessibility is still insufficient, and the overall picture of possible accessibility issues with SSTs is fragmented. In this study, we evaluated the accessibility of a sample of 20 SSTs in Japan by examining variables in user perception and action, factors related to cognitive accessibility, and user interface components. The findings are twofold. First, we illustrated the multimodalities in SST interaction based on the theory of human-computer interaction. Then, we identified SST user interface design practices that impact human cognition. This study illustrates the current reality of how accessibility is actualized and proposes future research directions and practices for the SST industry to develop and improve SSTs’ accessibility.

随着越来越多的服务只能以数字形式提供，自助服务技术（SSTs）需要易于获取，以确保所有公民都有平等的机会参与社会。然而，SSTs的可达性仍然不足，并且SSTs可能存在的可达性问题的整体情况是碎片化的。在这项研究中，我们通过考察用户感知和行为的变量、与认知可访问性相关的因素和用户界面组件，评估了日本20个sst样本的可访问性。研究结果是双重的。首先，基于人机交互理论，阐述了海表温度交互的多模态。然后，我们确定了影响人类认知的SST用户界面设计实践。本研究阐述了可达性实现的现状，并为海温行业发展和改善海温可达性提出了未来的研究方向和实践。

引用次数: 0