DroidCollection and DroidDetect provide the largest open resource for detecting AI-generated code, including adversarial 'humanized' samples, and show that training on a small amount of such data restores detector robustness.
In Proceedings of the 21st In- ternational Conference on Mining Software Reposito- ries, MSR ’24, page 394–406, New York, NY , USA
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SE 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
$\texttt{Droid}$: A Resource Suite for AI-Generated Code Detection
DroidCollection and DroidDetect provide the largest open resource for detecting AI-generated code, including adversarial 'humanized' samples, and show that training on a small amount of such data restores detector robustness.