20.500.12556/RUL-179046 Learning with noisy labels [~re]visited Learning with noisy labels (LNL) is a subfield of supervised machine learning investigating scenarios in which the training data contain errors. While most research has focused on synthetic noise, where labels are randomly corrupted, real-world noise from human annotation errors is more complex and less understood. Wei et al. (2022) introduced CIFAR-N, a dataset with human-labeled noise and claimed that real-world noise is fundamentally more challenging than synthetic noise. This study aims to reproduce their experiments on testing the characteristics of human-annotated label noise, memorization dynamics, and benchmarking of LNL methods. We successfully reproduce some of the claims but identify some quantitative discrepancies. Notably, our attempts to reproduce the reported benchmark reveal inconsistencies in the reported results. To address these issues, we develop a unified framework and propose a refined benchmarking protocol that ensures a fairer evaluation of LNL methods. Our findings confirm that real-world noise differs structurally from synthetic noise and is memorized more rapidly by deep networks. By open-sourcing our implementation, we provide a more reliable foundation for future research in LNL. noisy learning deep learning repeatability of experiments učenje na šumnih podatkih globoko učenje ponovljivost eksperimentov true false true Angleški jezik Slovenski jezik Članek v reviji 2026-02-04 08:44:19 2026-02-04 08:44:19 2026-02-05 04:27:09 0000-00-00 00:00:00 2026 0 0 22 str. iss. 1, Art. 2 Vol. 11 Jan. 2026 0000-00-00 Zaloznikova Objavljeno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 004.85 2430-3658 10.5281/zenodo.18401497 267165699 RAZ_Hudovernik_Valter_2026.pdf RAZ_Hudovernik_Valter_2026.pdf 1 25DDD9CA7FAADD0A5D7E0DA14F9CD556 5cbfc35d3fff7029a306bf5829ed5aacfb7e5106e356bc0889efaec49f514624 7c2af0d0-019e-11f1-a1ba-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=227148 https://rescience.github.io/bibliography/Hudovernik_2026.html 1 3ad31136-019d-11f1-a1ba-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=227147 https://zenodo.org/record/18401497/files/article.pdf 0 0de73136-019f-11f1-a1ba-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=227149 Fakulteta za računalništvo in informatiko 0 0 0