Аномалияны анықтаушы
method
ID: anomalydetector
Related Papers
Distinguishing Non-natural from Natural Adversarial Samples for More Robust Pre-trained Language Model
AbstractRecently, the problem of robustness of pre-trained language models (PrLMs) has received increasing research interest. Latest studies on adversarial attacks achieve high attack success rates against PrLMs, claiming that PrLMs …...