Replication Package of "Code and Architectural Debt in Artificial Intelligence-Enabled Systems: On theRelevance, Severity, Impact, and Mitigation Strategies"

Replication Package of the article "Code and Architectural Debt in Artificial Intelligence-Enabled Systems: On the<br>Relevance, Severity, Impact, and Mitigation Strategies".<br>In this paper, we leverage the expertise of practitioners to offer valuable insights to the research community, aiming to enhance awareness among researchers about the detection and mitigation of AI technical debt. Our ultimate goal is to empower practitioners by providing them with tools and methodologies. Additionally, our study sheds light on novel aspects that practitioners might not be fully acquainted with, contributing to a deeper understanding of the subject.We develop a survey study featuring 53 AI developers, in which we collect information on the practical relevance, severity, and impact of two key types of AI technical debt, such as architectural and code debt, other than the strategies applied by practitioners to identify and mitigate them.The key findings of the study reveal the multiple impacts that code and architectural debt may have on the quality of AI-enabled systems and the little support practitioners have to deal with them.We conclude the article by distilling lessons learned and actionable insights for researchers.

《人工智能赋能系统中的代码与架构债务：论相关性、严重性、影响及缓解策略》论文复现包。本研究依托行业从业者的专业经验，为学术共同体提供极具价值的研究视角，旨在提升研究者对人工智能技术债务（AI technical debt）的检测与缓解认知。本研究的终极目标是通过提供工具与方法论，赋能行业从业者。此外，本研究还揭示了从业者可能尚未充分了解的若干新兴维度，助力学界对该主题形成更深入的理解。我们开展了一项包含53名人工智能开发者的调研研究，收集了两类核心人工智能技术债务——架构债务与代码债务——的实际相关性、严重性及影响相关信息，同时采集了从业者用于识别与缓解此类债务的策略数据。本研究的核心发现揭示了代码债务与架构债务对人工智能赋能系统质量的多重负面影响，以及当前从业者在应对此类债务时可获得的支持极为有限。本文最后总结了研究所得的经验教训，并为研究者提供了可落地的实操启示。