Easy to Learn, Yet Hard to Forget: Towards Robust Unlearning Under Bias
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (7) · 5782-5790
DOI: 10.1609/aaai.v40i7.37499 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von JuneHyoung Kwon und 5 weiteren Autor:innen.
Abstract (Originalsprache)
Machine unlearning, which enables a model to forget specific data, is crucial for ensuring data privacy and model reliability. However, its effectiveness can be severely undermined in real-world scenarios where models learn unintended biases from spurious correlations within the data. This paper investigates the unique challenges of unlearning from such biased models. We identify a novel phenomenon we term "shortcut unlearning," where models exhibit an "easy to learn, yet hard to forget" tendency. Specifically, models struggle to forget easily-learned, bias-aligned samples; instead of forgetting the class attribute, they unlearn the bias attribute, which can paradoxically improve accuracy on the class intended to be forgotten. To address this, we propose CUPID, a new unlearning framework inspired by the observation that samples with different biases exhibit distinct loss landscape sharpness. Our method first partitions the forget set into causal- and bias-approximated subsets based on sample sharpness, then disentangles model parameters into causal and bias pathways, and finally performs a targeted update by routing refined causal and bias gradients to their respective pathways. Extensive experiments on biased datasets including Waterbirds, BAR, and Biased NICO++ demonstrate that our method achieves state-of-the-art forgetting performance and effectively mitigates the shortcut unlearning problem.
Zitieren
APA
Kwon, J., Kim, M., Lee, E., Lee, Y., Lee, S., & Kim, Y. (2026). Easy to Learn, Yet Hard to Forget: Towards Robust Unlearning Under Bias. Proceedings of the AAAI Conference on Artificial Intelligence, 40(7), 5782-5790. https://doi.org/10.1609/aaai.v40i7.37499
Vancouver
Kwon J, Kim M, Lee E, Lee Y, Lee S, Kim Y. Easy to Learn, Yet Hard to Forget: Towards Robust Unlearning Under Bias. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(7):5782-5790. doi:10.1609/aaai.v40i7.37499
BibTeX
@article{Kwon2026,
author = {Kwon, JuneHyoung and Kim, MiHyeon and Lee, Eunju and Lee, Yoonji and Lee, Seunghoon and Kim, YoungBin},
title = {Easy to Learn, Yet Hard to Forget: Towards Robust Unlearning Under Bias},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {7},
pages = {5782--5790},
doi = {10.1609/aaai.v40i7.37499}
}
EUVIMED