A Pilot Study on Detecting Software Design Patterns with Large Language Models: An Empirical Evaluation

Oishik Chowdhury, Bastin Tony Roy Savarimuthu and Sherlock A. Licorish School of Computing, University of Otago, New Zealand ([email protected], [email protected], [email protected]) · 2026

Read A Pilot Study on Detecting Software Design Patterns with Large Language Models: An Empirical Evaluation by Oishik Chowdhury, Bastin Tony Roy Savarimuthu and Sherlock A. Licorish School of Computing, University of Otago, New Zealand ([email protected], [email protected], [email protected]) (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-17329.txt.

Description

Chowdhury, Savarimuthu, and Licorish pilot-test whether large language models can spot classic software design patterns, such as singleton and adapter, in code. Four models and two ensembles were tried on code itself, diagram versions, and written descriptions, across five patterns. Early results favor NextCoder and Gemma 3, with ensembles lifting overall performance, suggesting language models could help newcomers grasp unfamiliar codebases and flag design problems.

Catalog record

CREATOROishik Chowdhury, Bastin Tony Roy Savarimuthu and Sherlock A. Licorish School of Computing, University of Otago, New Zealand ([email protected], [email protected], [email protected])
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE39671
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.17329

Browse the full library · About the Tower of Babel