week 58 - Evaluating LLMs for Automated Unit Test Maintenance, Specification-Driven Development and Test-Driven Code Generation with LLMs

A Socio-Technical Grounded Theory on the Effect of Cognitive Dysfunctions in the Performance of Software Developers with ADHD and Autism

The concept of neurodiversity, encompassing conditions such as Autism Spectrum Disorder (ASD), Attention-Deficit/Hyperactivity Disorder (ADHD), dyslexia, and dyspraxia, challenges traditional views of these neurodevelopmental variations as disorders and instead frames them as natural cognitive differences that contribute to unique ways of thinking and problem-solving. Within the software development industry, known for its emphasis on innovation, there is growing recognition of the value neurodivergent individuals bring to technical teams. Despite this, research on the contributions of neurodivergent individuals in Software Engineering (SE) remains limited. This interdisciplinary Socio-Technical Grounded Theory study addresses this gap by exploring the experiences of neurodivergent software engineers with ASD and ADHD, examining the cognitive and emotional challenges they face in software teams. Based on interviews and a survey with 25 neurodivergent and 5 neurotypical individuals, our theory describes how neurodivergent cognitive dysfunctions affect SE performance, and how the individuals' individual journey and various accommodations can regulate this effect. We conclude our paper with a list of inclusive Agile practices, allowing organizations to better support neurodivergent employees and fully leverage their capabilities.

See full paper - Last Accessed (Jan 30, 2026)

TAM-Eval: Evaluating LLMs for Automated Unit Test Maintenance

While Large Language Models (LLMs) have shown promise in software engineering, their application to unit testing remains largely confined to isolated test generation or oracle prediction, neglecting the broader challenge of test suite maintenance. We introduce TAM-Eval (Test Automated Maintenance Evaluation), a framework and benchmark designed to evaluate model performance across three core test maintenance scenarios: creation, repair, and updating of test suites. Unlike prior work limited to function-level tasks, TAM-Eval operates at the test file level, while maintaining access to full repository context during isolated evaluation, better reflecting real-world maintenance workflows. Our benchmark comprises 1,539 automatically extracted and validated scenarios from Python, Java, and Go projects. TAM-Eval supports system-agnostic evaluation of both raw LLMs and agentic workflows, using a reference-free protocol based on test suite pass rate, code coverage, and mutation testing. Empirical results indicate that state-of-the-art LLMs have limited capabilities in realistic test maintenance processes and yield only marginal improvements in test effectiveness. We release TAM-Eval as an open-source framework to support future research in automated software testing.

See full paper at arxiv.org - Last Accessed (Jan 30, 2026)

An Exploratory Study of Bayesian Prompt Optimization for Test-Driven Code Generation with Large Language Models

We consider the task of generating functionally correct code using large language models (LLMs). The correctness of generated code is influenced by the prompt used to query the given base LLM. We formulate the problem of finding the appropriate prompt as combinatorial search process and propose a Bayesian optimization (BO) approach referred to as {\em BO for Code GENeration (BODE-GEN)}. BODE-GEN performs an adaptive data-driven search over prompts guided by training data in the form of prompts tried and the functional accuracy of the generated code over a set of given test cases. The key insight is to perform BO in continuous embedding space by using an auxiliary LLM to bridge the gap between discrete prompt space and continuous embedding space. We leverage two synergistic ideas, namely, random projections and dimensionality scaled priors, to build effective Gaussian process based surrogate models over the high-dimensional embedding space. Our experiments on the HumanEval+ benchmark using multiple base LLMs show that BODE-GEN can improve performance in terms of code generation accuracy compared to fixed prompts and manual prompt engineering. Additionally, we demonstrate that BODE-GEN is sample-efficient, requiring relatively few iterations of BO to demonstrate improvements in code accuracy.

See full paper - Last Accessed (Jan 30, 2026)

Recommending Move Method Refactoring Opportunities Based on Feature Fusion and DeepLearning

The Move Method refactoring is crucial for mitigating the Feature Envy code smell, which enhances cohesion and reduces coupling by relocating methods to more suitable classes. Existing deep learning approaches often suffer from redundant features, limiting model generalization. To address this, this paper introduces GMove, a novel approach leveraging feature fusion and a hybrid deep learning architecture (Bi-LSTMand CNN branches) to recommend refactoring opportunities. By fusing semantic, structural, and metric features from a constructed 16,828-sample dataset, GMove effectively filters redundant information. Experimental results demonstrate that GMove achieves a high synthetic F1 score of 97.7% and significantly out performs state-of-the-art refactoring tools, showing an average F1 improvement of 9.7% over the strongest modern baseline, affirming its effectiveness and novel fusion strategy

See full paper - Last Accessed (Feb 2, 2026)

Agile Specification-Driven Development

We present an agile approach to Specification-Driven Development, which combines features of Test-Driven Development and the plan-based approach of Design-by-Contract. We argue that both tests and contracts are different types of specifications, and both are useful and complementary for building high quality software. We conclude that it is useful for being able to switch between writing tests and writing contracts, and explain how Specification-Driven Development supports this capability.

See full paper - Last Accessed (Feb 2, 2026)