Are We Testing or Being Tested? Exploring the Practical Applications of Large Language Models in Software Testing
Bibliographic record
Abstract
A Large Language Model (LLM) represents a cutting-edge artificial intelligence model that generates content, including grammatical sentences, human-like paragraphs, and syntactically code snippets. LLMs can play a pivotal role in soft-ware development, including software testing. LLMs go beyond traditional roles such as requirement analysis and documentation and can support test case generation, making them valuable tools that significantly enhance testing practices within the field. Hence, we explore the practical application of LLMs in software testing within an industrial setting, focusing on their current use by professional testers. In this context, rather than relying on existing data, we conducted a cross-sectional survey and collected data within real working contexts-specifically, engaging with practitioners in industrial settings. We applied quantitative and qualitative techniques to analyze and synthesize our collected data. Our findings demonstrate that LLMs effectively enhance testing documents and significantly assist testing professionals in programming tasks like debugging and test case automation. LLMs can support individuals engaged in manual testing who need to code. However, it is crucial to emphasize that, at this early stage, software testing professionals should use LLMs with caution while well-defined methods and guidelines are being built for the secure adoption of these tools.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.039 | 0.115 |
| Meta-epidemiology (narrow) | 0.000 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.002 | 0.010 |
| Scholarly communication | 0.007 | 0.017 |
| Open science | 0.002 | 0.004 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.003 | 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".