Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2501.01873

Latent Mutants: A large-scale study on the Interplay between mutation testing and software evolution

Jeongju Sohn·Ezekiel Soremekun·Michail Papadakis
View original source ↗Read source document ↗Export metadata ↓

Abstract

In this paper we apply mutation testing in an in-time fashion, i.e., across multiple project releases. Thus, we investigate how the mutants of the current version behave in the future versions of the programs. We study the characteristics of what we call latent mutants, i.e., the mutants that are live in one version and killed in later revisions, and explore whether they are predictable with these properties. We examine 131,308 mutants generated by Pitest on 13 open-source projects. Around 11.2% of these mutants are live, and 3.5% of them are latent, manifesting in 104 days on average. Using the mutation operators and change-related features we successfully demonstrate that these latent mutants are identifiable, predicting them with an accuracy of 86% and a balanced accuracy of 67% using a simple random forest classifier.

Explore related subjects

cs.SE

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Jeongju Sohn, Ezekiel Soremekun, Michail Papadakis. 2025-01-03. Latent Mutants: A large-scale study on the Interplay between mutation testing and software evolution. https://arxiv.org/abs/2501.01873

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Is Vibe Coding Safe? Benchmarking Vulnerability of Agent-Generated Code in Real-World Tasks

Vibe coding is a new software development paradigm in which human engineers prompt a large language model (LLM) agent to complete complex coding tasks with little supervision. Although vibe coding is increasingly adopted, is the generated code really safe to deploy in production? To investigate this question, we propose SUSVIBES, a benchmark consisting of 186 feature-request software engineering tasks from real-world open-source projects, for which, human programmers committed vulnerable implementations. We evaluate 12 widely used coding agentic settings with frontier models on the benchmark. Disturbingly, all agents perform poorly in terms of software security. Although 57% of the solutions from SWE-Agent with Claude 4 Sonnet are functionally correct, only 11.8% are secure. Further experiments demonstrate that preliminary security strategies, such as augmenting the feature request with vulnerability hints, cannot mitigate these security issues. Our findings raise serious concerns about the widespread adoption of vibe coding, particularly in security-sensitive applications. The code and dataset are available at https://github.com/LeiLiLab/susvibes. The leaderboard is at https://leililab.github.io/susvibes-leaderboard.

cs.SE↗
arXiv2026

Why3-py: A Tool for Formal Verification of Hypothesis Testing and Meta-Analysis in Python

The reproducibility crisis in scientific research has received widespread recognition, thereby increasing the importance of meta-analyses that integrate statistical analyses from multiple studies. However, statistical methods often have ambiguous and implicit underlying assumptions, which can lead to their erroneous applications and interpretations. To address this issue, we propose a formal verification framework for statistical Python programs. Specifically, we present Why3-py, a Python front-end for the Why3 verification platform that transforms Python code into verification-oriented WhyML representations, addressing the challenges arising from Python's dynamic typing and runtime polymorphism. Furthermore, we extend the StatWhy tool to support the verification of meta-analysis methods. These tools enable meta-analysts to identify overlooked assumptions and misuse of analyses, and to verify the correct use of hypothesis testing and meta-analysis methods in Python code.

cs.SE↗
arXiv2026

What is the Difference Between Me and You? Benchmarking the Quality Gap Between Human-Written and AI-Generated Code

AI coding assistants are becoming co-authors of production software, yet their evaluation centers on functional correctness, leaving open whether their code differs from human code in the quality dimensions dominating lifecycle cost. We compare human-written and AI-generated code at scale: 787,562 function pairs across Python, Java, and C, each human function mined from open-source repositories paired with implementations generated from its docstring by three AI assistants (OpenAI GPT models, DeepSeek-Coder, Qwen2.5-Coder). We characterize structural complexity and statistical naturalness, and map static-analysis findings onto Orthogonal Defect Classification for defects and the Common Weakness Enumeration for vulnerabilities, making authors and languages directly comparable. AI-generated code is structurally compressed and stylistically templated: roughly half the size and branching of human code, clustering apart at the style level. Defect profiles differ in kind: human code concentrates issues of mature codebases, AI code repetitive boilerplate; security is language-dependent, with LLMs producing more, and more severe, findings in Python and Java but fewer high-severity memory-safety findings than humans in C. Once size is controlled for, complexity metrics carry little signal, while naturalness separates authors. Finally, we release CQBench, a benchmark of 27,346 issue-prone tasks with baselines and an evaluation pipeline for quality assurance and security testing.

cs.SE↗