MuMonDE: A framework for evaluating model clone detectors using model mutation analysis
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Summary Model‐driven engineering is an increasingly prevalent approach in software engineering where models are the primary artifacts throughout a project's life cycle. A growing form of analysis and quality assurance in these projects is model clone detection, which identifies similar model elements. As model clone detection research and tools emerge, methods must be established to assess model clone detectors and techniques. In this paper, we describe the MuMonDE framework, which researchers and practitioners can use to evaluate model clone detectors using mutation analysis on the models each detector is geared towards. MuMonDE applies mutation testing in a novel way by randomly mutating model elements within existing projects to emulate various types of clones that can exist within that domain. It consists of 2 main phases. The mutation phase involves determining the mutation targets, selecting the appropriate mutation operations, and injecting mutants. The second phase, evaluation, involves detecting model clones, preprocessing clone reports, analyzing those reports to calculate recall and precision, and visualizing the data. We introduce MuMonDE by describing each phase in detail. We present our experiences and examples in successfully developing a MuMonDE implementation capable of evaluating Simulink model clone detectors. We validate MuMonDE by demonstrating its ability to answer evaluation questions and provide insights based on the data it generates. With this research using mutation analysis, our goal is to improve model clone detection and its analytical capabilities, thus improving model‐driven engineering as a whole.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.029 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it