Article

From Ambiguous Requirements to Hallucinated Tests: A Multi-Domain Study of the Mirror Effect in LLM-Assisted Software Testing

2026-08-21

Abstract excerpt

Although large language models (LLMs) have the capacity to produce test cases rapidly, the accuracy of the test cases relies heavily on the specification given to the model. In this paper, we present Mirror Effect: an LLM is likely to reflect any level of clarity or ambiguity in natural-language requirements in subsequent test artifacts. For a comparative within-subjects experiment with five expert QA engineers, t...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: fed42021-351f-5178-bb5a-84d72cf29b68
DOI: 10.20944/preprints202608.1538.v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

From Ambiguous Requirements to Hallucinated Tests: A Multi-Domain Study of the Mirror Effect in LLM-Assisted Software TestingDOI 10.20944/preprints202608.1538.v1Read

LLM-Based Test Case Generation from Natural-Language Requirements: A Verified Multi-Domain Empirical Study with Symbolic Mutation IndicatorsSemantic proximity Read

Cross-Domain Tuple and Random Collision GuidedApproach for Creative Requirement GenerationUsing LLMSemantic proximity Read

The Role of Explainable AI in Automated Software Testing: Opportunities and ChallengesSemantic proximity Read

An Experiment with Focus on Security Through Large- Language Models Using Behavior-Driven DevelopmentSemantic proximity Read

When Generative AI Writes Test Cases: Scenario-Driven Evaluation of Generated TestsSemantic proximity Read

Optimizing GPT-Based Distractor Generation for the Korean CSAT English ExamSemantic proximity Read

Unit Test Case Generation using Large Language Models for Students in Computer ProgrammingSemantic proximity Read

Students-Centric Evaluation Survey for Exploring the Impact of LLMs on UML ModelingSemantic proximity Read

AI-Driven Code Documentation: Comparative Evaluation of LLMs for Commit Message GenerationSemantic proximity Read

Who Broke the Pipeline? Traceability andAccountability in Multi-Agent LLM SoftwareEngineering PipelinesSemantic proximity Read

AiDER: Auditing and Document Evaluation via Rule Compilation and Small Language Models - An Education Case StudySemantic proximity Read

A Systematic Literature Review on Generative AI in Software Engineering: Code Generation and RefactoringSemantic proximity Read