Content Diversity-guided Ambiguity Mitigation for Open-Set Noisy Label Learning
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (16) · 13889-13897
DOI: 10.1609/aaai.v40i16.38398 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Zhihao Zhou und 2 weiteren Autor:innen.
Abstract (Originalsprache)
Open-set noisy label learning faces a critical challenge in maintaining robust DNN performance when training data contain both in-distribution noisy (IDN) and out-of-distribution (OOD) samples. These noisy samples induce overconfident but erroneous predictions due to their ambiguous positions relative to category boundaries. Current methods address this by filtering noisy samples based on visual features alone, they fail to resolve the semantic ambiguity near decision boundaries, where limited visual cues lead to unreliable sample purification. To this end, we propose Content Diversity-guided Ambiguity Mitigation (CDgAM), a novel framework that leverages diverse contents to mitigate visual ambiguity in open-set noisy label learning. CDgAM leverages textual descriptions of intra-class commonality and inter-class disparity to dynamically refine semantic boundaries, reducing bias in prototype learning. To further suppress early-stage uncertainty in visual representations, we design a region-sensitive distillation regularization that transfers boundary-aware knowledge from a multimodal large language model to the target DNN. Extensive experiments conducted on various datasets with different noise levels demonstrate the effectiveness of our CDgAM, outperforming state-of-the-art methods for open-set noisy label learning.
Zitieren
APA
Zhou, Z., Li, R., & Li, X. (2026). Content Diversity-guided Ambiguity Mitigation for Open-Set Noisy Label Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 40(16), 13889-13897. https://doi.org/10.1609/aaai.v40i16.38398
Vancouver
Zhou Z, Li R, Li X. Content Diversity-guided Ambiguity Mitigation for Open-Set Noisy Label Learning. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(16):13889-13897. doi:10.1609/aaai.v40i16.38398
BibTeX
@article{Zhou2026,
author = {Zhou, Zhihao and Li, Rui and Li, Xueying},
title = {Content Diversity-guided Ambiguity Mitigation for Open-Set Noisy Label Learning},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {16},
pages = {13889--13897},
doi = {10.1609/aaai.v40i16.38398}
}
EUVIMED