BILAL ÖZTÜRK; KELVIN LLEINS ROJAS-CÓRDOVA. Same accuracy, different failures: Robustness of pixel-based, gradient-feature and augmented classifiers to common image corruptions. Journal of Artificial Intelligence,Machine Learning and Neural Network , [S. l.], v. 6, n. 2, p. 32–43, 2026. DOI: 10.55529/jaimlnn.62.32.43. Disponível em: https://journal.hmjournals.com/index.php/JAIMLNN/article/view/6667. Acesso em: 24 oct. 2026.