Who Tests the Tests? Model-Based Validation of AI-Generated Test Cases
Eduard Paul Enoiu
A conceptual framework that uses model-based testing to check AI-generated test cases and classify them as model-backed, model-extending or model-inconsistent.
Publications
Papers, deliverables and project documents from the MONA LISA consortium.
Test Design and Review Argumentation in AI-Assisted Test Generation
Eduard Paul Enoiu, Robert Feldt
A taxonomy and template that describe each AI-generated test case by its goal, claim, reason and evidence, supporting both test design and review.
(Over)Reliance on Test Agents in AI-Assisted Software Testing
Eduard Paul Enoiu
A study of overreliance on AI test agents as both an agency and an assurance problem, with a framework for collecting data on it in test agent workflows.
Benchmarking Large Language Models for Root Cause Analysis in Train Control Software Testing
Rahmanu Hermawan, Alessio Bucaioni, Eduard Paul Enoiu, Wasif Afzal
A benchmark of seven large language models for automated root cause analysis of test failures, using execution logs from Alstom’s digital twin-based testing of train control software.