LLMs like NextCoder and Gemma 3 show promise in detecting patterns such as singleton, adapter, bridge, composite, and decorator, with ensembles improving overall performance across the three input formats tested.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SE 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A Pilot Study on Detecting Software Design Patterns with Large Language Models: An Empirical Evaluation
LLMs like NextCoder and Gemma 3 show promise in detecting patterns such as singleton, adapter, bridge, composite, and decorator, with ensembles improving overall performance across the three input formats tested.