Searcharxiv⌕ Search

arXiv subjects

Robert Feldt

Publications and source records attributed to Robert Feldt.

At least 37 records · Page 2Linked to original sources

AI for Requirements Engineering: Industry adoption and Practitioner perspectives

The integration of AI for Requirements Engineering (RE) presents significant benefits but also poses real challenges. Although RE is fundamental to software engineering, limited research has examined AI adoption in RE. We surveyed 55 software practitioners to map AI usage across four RE phases: Elicitation, Analysis, Specification, and Validation, and four approaches for decision making: human-only decisions, AI validation, Human AI Collaboration (HAIC), and full AI automation. Participants also shared their perceptions, challenges, and opportunities when applying AI for RE tasks. Our data show that 58.2% of respondents already use AI in RE, and 69.1% view its impact as positive or very positive. HAIC dominates practice, accounting for 54.4% of all RE techniques, while full AI automation remains minimal at 5.4%. Passive AI validation (4.4 to 6.2%) lags even further behind, indicating that practitioners value AI's active support over passive oversight. These findings suggest that AI is most effective when positioned as a collaborative partner rather than a replacement for human expertise. It also highlights the need for RE-specific HAIC frameworks along with robust and responsible AI governance as AI adoption in RE grows.

cs.SE↗

Large Language Models in Thematic Analysis: Prompt Engineering, Evaluation, and Guidelines for Qualitative Software Engineering Research

As artificial intelligence advances, large language models (LLMs) are entering qualitative research workflows, yet no reproducible methods exist for integrating them into established approaches like thematic analysis (TA), one of the most common qualitative methods in software engineering research. Moreover, existing studies lack systematic evaluation of LLM-generated qualitative outputs against established quality criteria. We designed and iteratively refined prompts for Phases 2-5 of Braun and Clarke's reflexive TA, then tested outputs from multiple LLMs against codes and themes produced by experienced researchers. Using 15 interviews on software engineers' well-being, we conducted blind evaluations with four expert evaluators who applied rubrics derived directly from Braun and Clarke's quality criteria. Evaluators preferred LLM-generated codes 61% of the time, finding them analytically useful for answering the research question. However, evaluators also identified limitations: LLMs fragmented data unnecessarily, missed latent interpretations, and sometimes produced themes with unclear boundaries. Our contributions are threefold. First, a reproducible approach integrating refined, documented prompts with an evaluation framework to operationalize Braun and Clarke's reflexive TA. Second, an empirical comparison of LLM- and human-generated codes and themes in software engineering data. Third, guidelines for integrating LLMs into qualitative analysis while preserving methodological rigour, clarifying when and how LLMs can assist effectively and when human interpretation remains essential.

cs.SE↗

Challenges in Testing Large Language Model Based Software: A Faceted Taxonomy

Large Language Models (LLMs) and Multi-Agent LLMs (MALLMs) introduce non-determinism unlike traditional or machine learning software, requiring new approaches to verifying correctness beyond simple output comparisons or statistical accuracy over test datasets. This paper presents a taxonomy for LLM test case design, informed by research literature and our experience. Each facet is exemplified, and we conduct an LLM-assisted analysis of six open-source testing frameworks, perform a sensitivity study of an agent-based system across different model configurations, and provide working examples contrasting atomic and aggregated test cases. We identify key variation points that impact test correctness and highlight open challenges that the research, industry, and open-source communities must address as LLMs become integral to software systems. Our taxonomy defines four facets of LLM test case design, addressing ambiguity in both inputs and outputs while establishing best practices. It distinguishes variability in goals, the system under test, and inputs, and introduces two key oracle types: atomic and aggregated. Our findings reveal that current tools treat test executions as isolated events, lack explicit aggregation mechanisms, and inadequately capture variability across model versions, configurations, and repeated runs. This highlights the need for viewing correctness as a distribution of outcomes rather than a binary property, requiring closer collaboration between academia and practitioners to establish mature, variability-aware testing methodologies.

cs.SE↗

SETBVE: Quality-Diversity Driven Exploration of Software Boundary Behaviors

Software systems exhibit distinct behaviors based on input characteristics, and failures often occur at the boundaries between input domains. Traditional Boundary Value Analysis (BVA) relies on manual heuristics, while automated Boundary Value Exploration (BVE) methods typically optimize a single quality metric, risking a narrow and incomplete survey of boundary behaviors. We introduce SETBVE, a customizable, modular framework for automated black-box BVE that leverages Quality-Diversity (QD) optimization to systematically uncover and refine a broader spectrum of boundaries. SETBVE maintains an archive of boundary pairs organized by input- and output-based behavioral descriptors. It steers exploration toward underrepresented regions while preserving high-quality boundary pairs and applies local search to refine candidate boundaries. In experiments with ten integer-based functions, SETBVE outperforms the baseline in diversity, boosting archive coverage by 37 to 82 percentage points. A qualitative analysis reveals that SETBVE identifies boundary candidates the baseline misses. While the baseline method typically plateaus in both diversity and quality after 30 seconds, SETBVE continues to improve in 600-second runs, demonstrating better scalability. Even the simplest SETBVE configurations perform well in identifying diverse boundary behaviors. Our findings indicate that balancing quality with behavioral diversity can help identify more software edge-case behaviors than quality-focused approaches.

cs.SE↗

Cross-Functional AI Task Forces (X-FAITs) for AI Transformation of Software Organizations

This experience report introduces the Cross-Functional AI Task Force (X-FAIT) framework to bridge the gap between strategic AI ambitions and operational execution within software-intensive organizations. Drawing from an Action Research case study at a global Swedish enterprise, we identify and address critical barriers such as departmental fragmentation, regulatory constraints, and organizational inertia that can impede successful AI transformation. X-FAIT employs force field analysis, executive sponsorship, cross-functional integration, and systematic risk assessment strategies to coordinate efforts across organizational boundaries, facilitating knowledge sharing and ensuring AI initiatives align with objectives. The framework provides both theoretical insights into AI-driven organizational transformation and practical guidance for software organizations aiming to effectively integrate AI into their daily workflows and, longer-term, products.

cs.SE↗

An Empirical Study on Decision-Making Aspects in Responsible Software Engineering for AI

Incorporating responsible practices into software engineering (SE) for AI is essential to ensure ethical principles, societal impact, and accountability remain at the forefront of AI system design and deployment. This study investigates the ethical challenges and complexities inherent in responsible software engineering (RSE) for AI, underscoring the need for practical,scenario-driven operational guidelines. Given the complexity of AI and the relative inexperience of professionals in this rapidly evolving field, continuous learning and market adaptation are crucial. Through qualitative interviews with seven practitioners(conducted until saturation), quantitative surveys of 51 practitioners, and static validation of results with four industry experts in AI, this study explores how personal values, emerging roles, and awareness of AIs societal impact influence responsible decision-making in RSE for AI. A key finding is the gap between the current state of the art and actual practice in RSE for AI, particularly in the failure to operationalize ethical and responsible decision-making within the software engineering life cycle for AI. While ethical issues in RSE for AI largely mirror those found in broader SE process, the study highlights a distinct lack of operational frameworks and resources to guide RSE practices for AI effectively. The results reveal that current ethical guidelines are insufficiently implemented at the operational level, reinforcing the complexity of embedding ethics throughout the software engineering life cycle. The study concludes that interdisciplinary collaboration, H-shaped competencies(Ethical-Technical dual competence), and a strong organizational culture of ethics are critical for fostering RSE practices for AI, with a particular focus on transparency and accountability.

cs.SE↗

The Factors Influencing Well-Being in Software Engineers: A Cross-Country Mixed-Method Study

The well-being of software engineers is increasingly under strain due to the high-stress nature of their roles, which involve complex problem-solving, tight deadlines, and the pressures of rapidly evolving technologies. Despite increasing recognition of mental health challenges in software engineering, few studies focus on the factors that sustain or undermine well-being. Existing research often overlooks the interaction between personal, collaborative, and organisational influences on this unique population. This study fills this gap by investigating the specific factors affecting the well-being of software engineers. We conducted 15 qualitative interviews and complemented them with a confirmatory cross-country survey to validate and extend our findings to a broader population. Our mixed-methods approach provides a robust framework to identify key factors influencing well-being, including personal perceptions of well-being, interpersonal and collaborative dynamics, workplace support and recognition, organisational culture, and specific stressors inherent to software engineering. By offering a detailed, context-specific exploration of these factors, our study builds on existing literature and provides actionable insights for improving well-being in software engineering. We conclude with policy recommendations to inform organisational strategies and develop targeted interventions that address the specific challenges of this field, contributing to more sustainable and supportive work environments.

cs.SE↗

Capturing Semantic Flow of ML-based Systems

ML-based systems are software systems that incorporates machine learning components such as Deep Neural Networks (DNNs) or Large Language Models (LLMs). While such systems enable advanced features such as high performance computer vision, natural language processing, and code generation, their internal behaviour remain largely opaque to traditional dynamic analysis such as testing: existing analysis typically concern only what is observable from the outside, such as input similarity or class label changes. We propose semantic flow, a concept designed to capture the internal behaviour of ML-based system and to provide a platform for traditional dynamic analysis techniques to be adapted to. Semantic flow combines the idea of control flow with internal states taken from executions of ML-based systems, such as activation values of a specific layer in a DNN, or embeddings of LLM responses at a specific inference step of LLM agents. The resulting representation, summarised as semantic flow graphs, can capture internal decisions that are not explicitly represented in the traditional control flow of ML-based systems. We propose the idea of semantic flow, introduce two examples using a DNN and an LLM agent, and finally sketch its properties and how it can be used to adapt existing dynamic analysis techniques for use in ML-based software systems.

cs.SE↗

Adaptive Random Testing with Q-grams: The Illusion Comes True

Adaptive Random Testing (ART) has faced criticism, particularly for its computational inefficiency, as highlighted by Arcuri and Briand. Their analysis clarified how ART requires a quadratic number of distance computations as the number of test executions increases, which limits its scalability in scenarios requiring extensive testing to uncover faults. Simulation results support this, showing that the computational overhead of these distance calculations often outweighs ART's benefits. While various ART variants have attempted to reduce these costs, they frequently do so at the expense of fault detection, lack complexity guarantees, or are restricted to specific input types, such as numerical or discrete data. In this paper, we introduce a novel framework for adaptive random testing that replaces pairwise distance computations with a compact aggregation of past executions, such as counting the q-grams observed in previous runs. Test case selection then leverages this aggregated data to measure diversity (e.g., entropy of q-grams), allowing us to reduce the computational complexity from quadratic to linear. Experiments with a benchmark of six web applications, show that ART with q-grams covers, on average, 4x more unique targets than random testing, and 3.5x more than ART using traditional distance-based methods.

cs.SE↗

Qualitative Research Methods in Software Engineering: Past, Present, and Future

The paper entitled "Qualitative Methods in Empirical Studies of Software Engineering" by Carolyn Seaman was published in TSE in 1999. It has been chosen as one of the most influential papers from the third decade of TSE's 50 years history. In this retrospective, the authors discuss the evolution of the use of qualitative methods in software engineering research, the impact it's had on research and practice, and reflections on what is coming and deserves attention.

cs.SE↗

Automating a Complete Software Test Process Using LLMs: An Automotive Case Study

Vehicle API testing verifies whether the interactions between a vehicle's internal systems and external applications meet expectations, ensuring that users can access and control various vehicle functions and data. However, this task is inherently complex, requiring the alignment and coordination of API systems, communication protocols, and even vehicle simulation systems to develop valid test cases. In practical industrial scenarios, inconsistencies, ambiguities, and interdependencies across various documents and system specifications pose significant challenges. This paper presents a system designed for the automated testing of in-vehicle APIs. By clearly defining and segmenting the testing process, we enable Large Language Models (LLMs) to focus on specific tasks, ensuring a stable and controlled testing workflow. Experiments conducted on over 100 APIs demonstrate that our system effectively automates vehicle API testing. The results also confirm that LLMs can efficiently handle mundane tasks requiring human judgment, making them suitable for complete automation in similar industrial contexts.

cs.SE↗

Adaptive Testing for LLM-Based Applications: A Diversity-based Approach

The recent surge of building software systems powered by Large Language Models (LLMs) has led to the development of various testing frameworks, primarily focused on treating prompt templates as the unit of testing. Despite the significant costs associated with test input execution and output assessment, the curation of optimized test suites is yet overlooked in these tools, which calls for tailored test selection or prioritization strategies. In this paper, we show that diversity-based testing techniques, such as Adaptive Random Testing (ART) with appropriate string distance metrics, can be effectively applied to the testing of prompt templates. Our proposed adaptive testing approach adjusts the conventional ART process to this context by selecting new test inputs based on scores derived from existing test suite and their labelling results. Our results, obtained using various implementations that explore several string-based distances, confirm that our approach enables the discovery of failures with reduced testing budgets and promotes the generation of more varied outputs.

cs.SE↗

Human-Centered AI Transformation: Exploring Behavioral Dynamics in Software Engineering

As Artificial Intelligence (AI) becomes integral to software development, understanding the social and cooperative dynamics that affect AI-driven organizational change is important. Yet, despite AI's rapid progress and influence, the human and cooperative facets of these shifts in software organizations remain relatively less explored. This study uses Behavioral Software Engineering (BSE) as a lens to examine these often-overlooked dimensions of AI transformation. Through a qualitative approach involving ten semi-structured interviews across four organizations that are undergoing AI transformations, we performed a thematic analysis that revealed numerous sub-themes linked to twelve BSE concepts across individual, group, and organizational levels. Since the organizations are at an early stage of transformation we found more emphasis on the individual level. Our findings further reveal six key challenges tied to these BSE aspects that the organizations face during their AI transformation. Aligned with change management literature, we emphasize that effective communication, proactive leadership, and resistance management are essential for successful AI integration. However, we also identify ethical considerations as critical in the AI context-an area largely overlooked in previous research. Furthermore, a narrative analysis illustrates how different roles within an organization experience the AI transition in unique ways. These insights underscore that AI transformation extends beyond technical solutions; it requires a thoughtful approach that balances technological and human factors.

cs.SE↗

GoNoGo: An Efficient LLM-based Multi-Agent System for Streamlining Automotive Software Release Decision-Making

Traditional methods for making software deployment decisions in the automotive industry typically rely on manual analysis of tabular software test data. These methods often lead to higher costs and delays in the software release cycle due to their labor-intensive nature. Large Language Models (LLMs) present a promising solution to these challenges. However, their application generally demands multiple rounds of human-driven prompt engineering, which limits their practical deployment, particularly for industrial end-users who need reliable and efficient results. In this paper, we propose GoNoGo, an LLM agent system designed to streamline automotive software deployment while meeting both functional requirements and practical industrial constraints. Unlike previous systems, GoNoGo is specifically tailored to address domain-specific and risk-sensitive systems. We evaluate GoNoGo's performance across different task difficulties using zero-shot and few-shot examples taken from industrial practice. Our results show that GoNoGo achieves a 100% success rate for tasks up to Level 2 difficulty with 3-shot examples, and maintains high performance even for more complex tasks. We find that GoNoGo effectively automates decision-making for simpler tasks, significantly reducing the need for manual intervention. In summary, GoNoGo represents an efficient and user-friendly LLM-based solution currently employed in our industrial partner's company to assist with software release decision-making, supporting more informed and timely decisions in the release process for risk-sensitive vehicle systems.

cs.AI↗

Domain Generalization through Meta-Learning: A Survey

Deep neural networks (DNNs) have revolutionized artificial intelligence but often lack performance when faced with out-of-distribution (OOD) data, a common scenario due to the inevitable domain shifts in real-world applications. This limitation stems from the common assumption that training and testing data share the same distribution--an assumption frequently violated in practice. Despite their effectiveness with large amounts of data and computational power, DNNs struggle with distributional shifts and limited labeled data, leading to overfitting and poor generalization across various tasks and domains. Meta-learning presents a promising approach by employing algorithms that acquire transferable knowledge across various tasks for fast adaptation, eliminating the need to learn each task from scratch. This survey paper delves into the realm of meta-learning with a focus on its contribution to domain generalization. We first clarify the concept of meta-learning for domain generalization and introduce a novel taxonomy based on the feature extraction strategy and the classifier learning methodology, offering a granular view of methodologies. Additionally, we present a decision graph to assist readers in navigating the taxonomy based on data availability and domain shifts, enabling them to select and develop a proper model tailored to their specific problem requirements. Through an exhaustive review of existing methods and underlying theories, we map out the fundamentals of the field. Our survey provides practical insights and an informed discussion on promising research directions.

cs.LG↗

The Social Psychology of Software Security (Psycurity)

This position paper explores the intricate relationship between social psychology and secure software engineering, underscoring the vital role social psychology plays in the realm of engineering secure software systems. Beyond a mere technical endeavor, this paper contends that understanding and integrating social psychology principles into software processes are imperative for establishing robust and secure software systems. Recent studies in related fields show the importance of understanding the social psychology of other security domains. Finally, we identify critical gaps in software security research and present a set of research questions for incorporating more social psychology into software security research.

cs.SE↗

Towards Explainable Test Case Prioritisation with Learning-to-Rank Models

Test case prioritisation (TCP) is a critical task in regression testing to ensure quality as software evolves. Machine learning has become a common way to achieve it. In particular, learning-to-rank (LTR) algorithms provide an effective method of ordering and prioritising test cases. However, their use poses a challenge in terms of explainability, both globally at the model level and locally for particular results. Here, we present and discuss scenarios that require different explanations and how the particularities of TCP (multiple builds over time, test case and test suite variations, etc.) could influence them. We include a preliminary experiment to analyse the similarity of explanations, showing that they do not only vary depending on test case-specific predictions, but also on the relative ranks.

cs.SE↗

Semantic API Alignment: Linking High-level User Goals to APIs

Large Language Models (LLMs) are becoming key in automating and assisting various software development tasks, including text-based tasks in requirements engineering but also in coding. Typically, these models are used to automate small portions of existing tasks, but we present a broader vision to span multiple steps from requirements engineering to implementation using existing libraries. This approach, which we call Semantic API Alignment (SEAL), aims to bridge the gap between a user's high-level goals and the specific functions of one or more APIs. In this position paper, we propose a system architecture where a set of LLM-powered ``agents'' match such high-level objectives with appropriate API calls. This system could facilitate automated programming by finding matching links or, alternatively, explaining mismatches to guide manual intervention or further development. As an initial pilot, our paper demonstrates this concept by applying LLMs to Goal-Oriented Requirements Engineering (GORE), via sub-goal analysis, for aligning with REST API specifications, specifically through a case study involving a GitHub statistics API. We discuss the potential of our approach to enhance complex tasks in software development and requirements engineering and outline future directions for research.

cs.SE↗