Software Impacts最新文献

英文中文

Rosetta-XAI: An automated evaluation and explainability framework for code translation models Rosetta-XAI：用于代码翻译模型的自动评估和可解释性框架

IF 1.2 Q3 COMPUTER SCIENCE, SOFTWARE ENGINEERING

Software Impacts

Pub Date : 2026-01-19 DOI: 10.1016/j.simpa.2026.100811

Vishnu S. Pendyala, Neha Bais Thakur

This paper presents Rosetta-XAI, a comprehensive software framework for evaluating and explaining Large Language Model (LLM) behavior in cross-language code conversion tasks. The system implements a four-stage automated pipeline: (1) code generation by LLMs accessed through the Ollama API inference service, (2) regex-based extraction of code blocks from markdown responses, (3) language-specific syntax and compilation validation with temporary artifact management, and (4) execution with timeout protections and CSV-based checkpoint recovery. The framework supports evaluation of 15 specialized code LLMs (1.3B–34B parameters), including DeepSeek Coder, Code Llama, CodeGemma, and Granite Code across 17 Rosetta Code programming tasks, generating 42 bidirectional conversion pairs among seven languages (C, C++, Go, Java, JavaScript, Python, Rust). Beyond traditional pass@1 accuracy metrics, the system incorporates explainability analysis through Shapley Value Sampling and Feature Ablation techniques implemented via Captum and PyTorch, enabling researchers to quantify token-level feature importance during translation. All pipeline components include XAI-enhanced variants supporting follow-up question analysis for interpretability studies. Built using Python with pandas for metrics aggregation and subprocess management for multi-language execution, the modular architecture separates extraction, validation, and execution concerns. Results are systematically organized into structured directories tracking accepted code, compilation failures, syntax errors, and execution outputs, with comprehensive metrics exported to CSVs for reproducible research and comparative model analysis.

本文介绍了Rosetta-XAI，一个综合的软件框架，用于评估和解释跨语言代码转换任务中的大语言模型（LLM）行为。该系统实现了一个四阶段的自动化流水线：(1)通过Ollama API推理服务访问的llm生成代码；(2)从markdown响应中提取基于regex的代码块；(3)使用临时工件管理的特定语言语法和编译验证；(4)使用超时保护和基于csv的检查点恢复执行。该框架支持评估15个专门的代码llm （1.3 - 34b参数），包括DeepSeek Coder、code Llama、CodeGemma和Granite code，跨越17个Rosetta code编程任务，在7种语言（C、c++、Go、Java、JavaScript、Python、Rust）之间生成42个双向转换对。除了传统的pass@1精度指标外，该系统还通过Captum和PyTorch实现的Shapley值采样和特征消融技术结合了可解释性分析，使研究人员能够在翻译过程中量化标记级特征的重要性。所有管道组件都包括支持可解释性研究的后续问题分析的xai增强变体。使用Python和pandas构建，用于度量聚合和多语言执行的子流程管理，模块化体系结构将提取、验证和执行问题分开。结果被系统地组织到结构化目录中，跟踪可接受的代码、编译失败、语法错误和执行输出，并将综合指标导出到csv，用于可重复的研究和比较模型分析。

{"title":"Rosetta-XAI: An automated evaluation and explainability framework for code translation models","authors":"Vishnu S. Pendyala, Neha Bais Thakur","doi":"10.1016/j.simpa.2026.100811","DOIUrl":"10.1016/j.simpa.2026.100811","url":null,"abstract":"<div><div>This paper presents Rosetta-XAI, a comprehensive software framework for evaluating and explaining Large Language Model (LLM) behavior in cross-language code conversion tasks. The system implements a four-stage automated pipeline: (1) code generation by LLMs accessed through the Ollama API inference service, (2) regex-based extraction of code blocks from markdown responses, (3) language-specific syntax and compilation validation with temporary artifact management, and (4) execution with timeout protections and CSV-based checkpoint recovery. The framework supports evaluation of 15 specialized code LLMs (1.3B–34B parameters), including DeepSeek Coder, Code Llama, CodeGemma, and Granite Code across 17 Rosetta Code programming tasks, generating 42 bidirectional conversion pairs among seven languages (C, C++, Go, Java, JavaScript, Python, Rust). Beyond traditional pass@1 accuracy metrics, the system incorporates explainability analysis through Shapley Value Sampling and Feature Ablation techniques implemented via Captum and PyTorch, enabling researchers to quantify token-level feature importance during translation. All pipeline components include XAI-enhanced variants supporting follow-up question analysis for interpretability studies. Built using Python with pandas for metrics aggregation and subprocess management for multi-language execution, the modular architecture separates extraction, validation, and execution concerns. Results are systematically organized into structured directories tracking accepted code, compilation failures, syntax errors, and execution outputs, with comprehensive metrics exported to CSVs for reproducible research and comparative model analysis.</div></div>","PeriodicalId":29771,"journal":{"name":"Software Impacts","volume":"27 ","pages":"Article 100811"},"PeriodicalIF":1.2,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146023076","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An open-source reference framework for the implementation of type 3 Asset Administration Shells 一个用于实现类型3资产管理shell的开源参考框架

IF 1.2 Q3 COMPUTER SCIENCE, SOFTWARE ENGINEERING

Software Impacts

Pub Date : 2026-01-09 DOI: 10.1016/j.simpa.2025.100807

Ekaitz Hurtado , Isabel Sarachaga , Aintzane Armentia , Oskar Casquero

Seamlessly integrating assets into distributed digital ecosystems based on Industry 4.0/5.0 demands measurable impact: lower engineering cost, interoperability and adaptability. SMIA (Self-configurable Manufacturing Industrial Agents) addresses this as an reference framework for implementing autonomous digital counterparts of assets, unifying industrial and software standards. Its dual-layer architecture combines machine-interpretable semantic modeling with distributed functional software, enhancing interoperability, flexibility, and autonomy. SMIA represents assets by executing domain-specific tasks and performing peer-to-peer communication through standardized interfaces. Following open scientific software principles, it integrates mature technologies and provides reproducible deployment artifacts (e.g., Docker), ensuring traceability and extensibility while reducing engineering effort and technological fragmentation.

将资产无缝集成到基于工业4.0/5.0的分布式数字生态系统中需要可衡量的影响：更低的工程成本、互操作性和适应性。SMIA（自配置制造工业代理）解决了这个问题，作为实现资产的自主数字对等物的参考框架，统一了工业和软件标准。它的双层架构结合了机器可解释的语义建模和分布式功能软件，增强了互操作性、灵活性和自主性。SMIA通过执行特定于领域的任务和通过标准化接口执行点对点通信来表示资产。遵循开放的科学软件原则，它集成了成熟的技术，并提供可重复的部署工件（例如Docker），确保了可追溯性和可扩展性，同时减少了工程工作量和技术碎片。

引用次数: 0

MatAbaAutoRel: A MATLAB–Abaqus framework for automated reliability analysis MatAbaAutoRel：用于自动化可靠性分析的MATLAB-Abaqus框架

IF 1.2 Q3 COMPUTER SCIENCE, SOFTWARE ENGINEERING

Software Impacts

Pub Date : 2026-01-07 DOI: 10.1016/j.simpa.2025.100808

S. Malekpour , S.M. Dehghan , M.A. Najafgholipour , S. Behravesh

Probabilistic and reliability analyses utilizing finite element software are frequently constrained by manual model creation and result extraction. This study presents an open-source, MATLAB-based framework integrated with Abaqus that automates randomized model generation through Monte Carlo simulation, performs analyses, and retrieves target results via a lightweight Python script in noGUI mode. The modular tool reduces user intervention and facilitates automated variations in geometry, material properties, and loading conditions. This framework enables rapid model generation and result extraction for hundreds of analyses in seconds, significantly reducing manual effort and potential human error.

利用有限元软件进行概率和可靠性分析经常受到人工模型创建和结果提取的限制。本研究提出了一个基于matlab的开源框架，与Abaqus集成，通过蒙特卡罗模拟自动生成随机模型，执行分析，并通过noGUI模式下的轻量级Python脚本检索目标结果。模块化工具减少了用户的干预，并促进了几何形状、材料属性和加载条件的自动变化。该框架支持在几秒钟内快速生成模型和提取数百个分析的结果，显著减少了人工工作和潜在的人为错误。

引用次数: 0

ASReview Dory: Bringing new and exciting models to ASReview LAB ASReview多利：为ASReview LAB带来新的和令人兴奋的模型

IF 1.2 Q3 COMPUTER SCIENCE, SOFTWARE ENGINEERING

Software Impacts

Pub Date : 2026-01-06 DOI: 10.1016/j.simpa.2025.100809

Timo van der Kuil , Jelle Jasper Teijema , Jonathan de Bruin , Rens van de Schoot

Systematic reviewing is a time-consuming process which can be accelerated through screening prioritisation via active learning. ASReview Dory enables researchers to test, validate, and apply a wide range of embedders and classifiers in systematic literature screening. It extends ASReview LAB, an open source, lightweight, and user-friendly environment with proven default models and extensibility through Python entry points. ASReview Dory adds ready-to-use transformer-based embedders, neural classifiers, and a framework for integrating custom models. Once installed, these models are directly available in ASReview LAB without additional configuration and can be systematically evaluated using the API or ASReview Makita.

系统审查是一个耗时的过程，可以通过主动学习筛选优先级来加速。ASReview Dory使研究人员能够在系统的文献筛选中测试，验证和应用广泛的嵌入器和分类器。它扩展了ASReview LAB， ASReview LAB是一个开源的、轻量级的、用户友好的环境，具有经过验证的默认模型和通过Python入口点的可扩展性。ASReview Dory添加了现成的基于变压器的嵌入器、神经分类器和集成自定义模型的框架。一旦安装，这些模型就可以直接在ASReview LAB中使用，而无需额外的配置，并且可以使用API或ASReview Makita系统地进行评估。

引用次数: 0

BEARING-FDD: An early detection and diagnosis tool for bearing faults in rotating machinery 轴承fdd：一种旋转机械轴承故障的早期检测和诊断工具

IF 1.2 Q3 COMPUTER SCIENCE, SOFTWARE ENGINEERING

Software Impacts

Pub Date : 2026-01-03 DOI: 10.1016/j.simpa.2025.100810

L. Magadán , C. Ruiz-Cárcel , J.C. Granda , F.J. Suárez , A. Menéndez-González , A. Starr

This paper presents the design and implementation of a web tool offering an innovative method for detecting, diagnosing and classifying bearing faults in rotating machinery under limited data conditions, providing explainability and interpretability of the results obtained. The tool uses a machine learning model to detect and diagnose bearing faults. A monotonic smoothed stacked autoencoder builds a health indicator without requiring feature extraction, making the tool useful without the need for specialized staff. The tool generates explainability and interpretability reports with a correlation analysis between the health indicator and well-known engineering features and easily interpretable details on the diagnosed faults. The tool includes the option to use preloaded state-of-the-art datasets, while also allowing users to upload their own datasets to analyze vibration data from real industrial equipment.

本文介绍了一个web工具的设计和实现，该工具提供了一种在有限数据条件下检测、诊断和分类旋转机械轴承故障的创新方法，并提供了结果的可解释性和可解释性。该工具使用机器学习模型来检测和诊断轴承故障。单调平滑堆叠自编码器在不需要特征提取的情况下构建运行状况指示器，使该工具无需专业人员即可使用。该工具生成可解释性和可解释性报告，其中包含运行状况指标与众所周知的工程特征之间的相关性分析，以及诊断故障的易于解释的详细信息。该工具包括使用预加载的最先进数据集的选项，同时还允许用户上传自己的数据集，以分析来自实际工业设备的振动数据。

引用次数: 0

SMCS: A lightweight MobileNet-based framework for skin cancer classification, segmentation, and explanation SMCS：一个轻量级的基于mobilenet的皮肤癌分类、分割和解释框架

IF 1.2 Q3 COMPUTER SCIENCE, SOFTWARE ENGINEERING

Software Impacts

Pub Date : 2025-12-15 DOI: 10.1016/j.simpa.2025.100804

Thuan Van Tran, Triet Minh Nguyen, Quy Thanh Lu

Skin is an important part of the guardian system, which helps to protect us from harmful factors such as physical impact, bacteria, viruses, and especially daily ultraviolet (UV) radiation. However, the changing of the environment in the present era creates prolonged exposure to UV, which can damage the skin and increase the risk of skin cancer. Thus, a skin cancer classification and detection framework called SMCS (Sampling in MobileNet for Skin Classification) was published by taking the power of artificial intelligence and deep learning. In this pipeline, skin illnesses can be discovered early, which aids doctors and patients in diagnosis and treatment while reducing both time and cost.

皮肤是保护系统的重要组成部分，它有助于保护我们免受有害因素的影响，如物理冲击，细菌，病毒，尤其是日常紫外线（UV）辐射。然而，当今时代环境的变化使人长时间暴露在紫外线下，这会损害皮肤，增加患皮肤癌的风险。因此，利用人工智能和深度学习的力量，发表了一个名为SMCS （Sampling in MobileNet for skin classification）的皮肤癌分类检测框架。在这个管道中，皮肤疾病可以早期发现，这有助于医生和患者的诊断和治疗，同时减少时间和成本。

引用次数: 0

Lcpy: An open-source python package for parametric and dynamic life cycle assessment and life cycle costing analysis Lcpy：一个用于参数化和动态生命周期评估以及生命周期成本分析的开源python包

IF 1.2 Q3 COMPUTER SCIENCE, SOFTWARE ENGINEERING

Software Impacts

Pub Date : 2025-12-05 DOI: 10.1016/j.simpa.2025.100805

Spiros Gkousis, Evina Katsou

Life Cycle Assessment (LCA) and Life Cycle Costing (LCC) are becoming key methods for sustainability analysis. Current software solutions usually focus on one method, omitting synergies and the provision of a holistic picture of system sustainability. Integrating LCA and LCC software with complex system models, uncertainty, and optimization tools remains a barrier for integrated techno-sustainability assessments. Lcpy is an open-source python package that enables using parametric or simulation process models, in-time projections, multiple scenarios, and flexible modelling for simple and dynamic LCA and LCC, uncertainty analysis and optimization. Visualization and storage functions are provided allowing end-to-end LCA and LCC analyses.

生命周期评价（LCA）和生命周期成本计算（LCC）正在成为可持续发展分析的关键方法。当前的软件解决方案通常集中在一种方法上，忽略了协同作用和提供系统可持续性的整体画面。将LCA和LCC软件与复杂的系统模型、不确定性和优化工具集成仍然是集成技术可持续性评估的障碍。Lcpy是一个开源python包，可以使用参数化或仿真过程模型、实时预测、多场景和灵活建模，用于简单和动态的LCA和LCC、不确定性分析和优化。提供可视化和存储功能，允许端到端的LCA和LCC分析。

引用次数: 0

GO-HKP: A Gene Ontology hierarchy-driven framework for enzyme kcat prediction GO-HKP：用于酶kcat预测的基因本体层次驱动框架

IF 1.2 Q3 COMPUTER SCIENCE, SOFTWARE ENGINEERING

Software Impacts

Pub Date : 2025-12-01 DOI: 10.1016/j.simpa.2025.100803

Qinghan Meng , Zhitao Mao , Hao Chen , Yuanyuan Huang , Hongwu Ma

GO-HKP is a Gene Ontology hierarchy-driven framework for predicting enzyme turnover numbers (

k_{cat}

) with improved coverage, generalizability, and interpretability. It integrates curated UniProt data, ontology-based

k_{cat}

propagation, and sequence-driven GO annotation (DeepGO-SE) to infer

k_{cat}

for both annotated and novel enzymes. Benchmarking across four genome-scale metabolic models demonstrated substantial improvements in reaction coverage — by 56.67%, 25.1%, 16.0%, and 14.5% — compared with existing methods, highlighting its strong gap-filling capability. GO-HKP offers a biologically grounded, scalable, and transparent approach, supporting applications in metabolic engineering, drug discovery, and systems biology. The framework and Python package are available via GitHub for broad usability and reproducibility.

GO-HKP是一个基因本体论层次驱动的框架，用于预测酶周转数（kcat），具有更好的覆盖率、通用性和可解释性。它集成了整理的UniProt数据、基于本体的kcat传播和序列驱动的GO注释（DeepGO-SE），以推断已注释酶和新酶的kcat。四种基因组尺度代谢模型的基准测试表明，与现有方法相比，反应覆盖率有显著提高——分别提高56.67%、25.1%、16.0%和14.5%，突出了其强大的空白填补能力。GO-HKP提供生物学基础，可扩展和透明的方法，支持代谢工程，药物发现和系统生物学的应用。该框架和Python包可通过GitHub获得，具有广泛的可用性和可重复性。

引用次数: 0

Software architecture description in original software publications 原始软件出版物中的软件体系结构描述

IF 1.2 Q3 COMPUTER SCIENCE, SOFTWARE ENGINEERING

Software Impacts

Pub Date : 2025-11-28 DOI: 10.1016/j.simpa.2025.100802

Tomasz Górski

A software architecture description is a work product that reveals software architecture. An architecture view manifests the system architecture from a specific perspective. In publications presenting original software, it is crucial to introduce the functions implemented by the software and identify its users. The structure and operation of the software should also be depicted. However, many publications contain drawings that often combine content from several views. Therefore, the paper introduces a method for describing software architecture in Use Cases and Logical views of the 1+5 model. The method expresses the architecture of a new software package for real estate sales.

软件体系结构描述是揭示软件体系结构的工作产品。架构视图从一个特定的角度显示系统架构。在介绍原始软件的出版物中，介绍软件实现的功能和识别其用户是至关重要的。还应描述软件的结构和操作。然而，许多出版物包含的绘图通常结合了来自几个视图的内容。因此，本文介绍了一种在用例和1+5模型的逻辑视图中描述软件体系结构的方法。该方法表达了一个新的房地产销售软件包的体系结构。

引用次数: 0

‘Forensic-DataFusion-Tool’: A Python-based application for exploratory forensic data analysis using merged datasets from analytical sensors “法医数据融合工具”：一个基于python的应用程序，用于探索性法医数据分析，使用来自分析传感器的合并数据集

IF 1.2 Q3 COMPUTER SCIENCE, SOFTWARE ENGINEERING

Software Impacts

Pub Date : 2025-10-01 DOI: 10.1016/j.simpa.2025.100799

Giorgio Felizzato , Michele Verdi , Angelo Michele Gargantini , Nico Pellegrinelli , Francesco Saverio Romolo

Portable sensors for on-site forensic analysis have advanced significantly, enabling reliable methods for crime scene investigation. Non-destructive analytical instruments are especially useful for providing chemical information from the same specimen. Combining data from these instruments through data fusion enhances analytical responses. Data fusion merges data from different sources to improve exploratory and predictive models. No current application supports multi-dataset fusion on a single platform. To address this, we developed a Python-based ‘Forensic-DataFusion-Tool’ to merge raw and preprocessed data from multiple sensors, speeding up data fusion and enabling future machine learning updates, including classification algorithms.

用于现场法医分析的便携式传感器取得了重大进展，为犯罪现场调查提供了可靠的方法。非破坏性分析仪器对于提供同一样品的化学信息特别有用。通过数据融合将来自这些仪器的数据结合起来，可以增强分析响应。数据融合将来自不同来源的数据合并在一起，以改进探索性和预测性模型。目前没有应用程序支持单一平台上的多数据集融合。为了解决这个问题，我们开发了一个基于python的“取证-数据融合工具”，用于合并来自多个传感器的原始和预处理数据，加速数据融合，并使未来的机器学习更新成为可能，包括分类算法。

引用次数: 0

下一页尾页

类型

全部化学•材料生命科学医学物理工程技术环境•农林材料科学地球科学法学管理学化学环境科学与生态学计算机科学教育学经济学农林科学人文科学生物学数学物理与天体物理心理学综合性期刊其他工业工程理学历史学农学文学信息工程

数据库

全部 ACS Publications Elsevier ieeexplore Springer The Royal Society of Chemistry Wiley

期刊

Software Impacts

全部 Acc. Chem. Res. ACS Applied Bio Materials ACS Appl. Electron. Mater. ACS Appl. Energy Mater. ACS Appl. Mater. Interfaces ACS Appl. Nano Mater. ACS Appl. Polym. Mater. ACS BIOMATER-SCI ENG ACS Catal. ACS Cent. Sci. ACS Chem. Biol. ACS Chemical Health & Safety ACS Chem. Neurosci. ACS Comb. Sci. ACS Earth Space Chem. ACS Energy Lett. ACS Infect. Dis. ACS Macro Lett. ACS Mater. Lett. ACS Med. Chem. Lett. ACS Nano ACS Omega ACS Photonics ACS Sens. ACS Sustainable Chem. Eng. ACS Synth. Biol. Anal. Chem. BIOCHEMISTRY-US Bioconjugate Chem. BIOMACROMOLECULES Chem. Res. Toxicol. Chem. Rev. Chem. Mater. CRYST GROWTH DES ENERG FUEL Environ. Sci. Technol. Environ. Sci. Technol. Lett. Eur. J. Inorg. Chem. IND ENG CHEM RES Inorg. Chem. J. Agric. Food. Chem. J. Chem. Eng. Data J. Chem. Educ. J. Chem. Inf. Model. J. Chem. Theory Comput. J. Med. Chem. J. Nat. Prod. J PROTEOME RES J. Am. Chem. Soc. LANGMUIR MACROMOLECULES Mol. Pharmaceutics Nano Lett. Org. Lett. ORG PROCESS RES DEV ORGANOMETALLICS J. Org. Chem. J. Phys. Chem. J. Phys. Chem. A J. Phys. Chem. B J. Phys. Chem. C J. Phys. Chem. Lett. Analyst Anal. Methods Biomater. Sci. Catal. Sci. Technol. Chem. Commun. Chem. Soc. Rev. CHEM EDUC RES PRACT CRYSTENGCOMM Dalton Trans. Energy Environ. Sci. ENVIRON SCI-NANO ENVIRON SCI-PROC IMP ENVIRON SCI-WAT RES Faraday Discuss. Food Funct. Green Chem. Inorg. Chem. Front. Integr. Biol. J. Anal. At. Spectrom. J. Mater. Chem. A J. Mater. Chem. B J. Mater. Chem. C Lab Chip Mater. Chem. Front. Mater. Horiz. MEDCHEMCOMM Metallomics Mol. Biosyst. Mol. Syst. Des. Eng. Nanoscale Nanoscale Horiz. Nat. Prod. Rep. New J. Chem. Org. Biomol. Chem. Org. Chem. Front. PHOTOCH PHOTOBIO SCI PCCP Polym. Chem.

﹀