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.
Bias Redistribution in Visual Machine Unlearning: Does Forgetting One Group Harm Another?
NewraLab；Beihang University（北京航空航天大学）；GoerTek Alpha Lab（歌尔Alpha实验室）；Skyline University Nigeria（尼日利亚天际线大学）；UESTC（电子科技大学）；Imperial College London（帝国理工学院）
Forget-It-All: Multi-Concept Machine Unlearning via Concept-Aware Neuron Masking
The University of Arizona（亚利桑那大学）；The University of Tulsa（塔尔萨大学）；Clemson University（克莱姆森大学）；Western Digital Corporation（西部数据公司）；University of Georgia（佐治亚大学）
Are We Truly Forgetting? A Critical Re-examination of Machine Unlearning Evaluation Protocols
Department of Artificial Intelligence, Korea University, South Korea（韩国大学人工智能系）；Computer Science Dept. at Courant Institute of Mathemtical Sciences, New York University, United States（纽约大学Courant数学科学研究所计算机科学系）
Forget by Uncertainty: Orthogonal Entropy Unlearning for Quantized Neural Networks
Hubei Key Laboratory of Transportation Internet of Things, School of Computer Science and Artificial Intelligence, Wuhan University of Technology, China（交通物联网湖北重点实验室，计算机科学与人工智能学院，武汉理工大学，中国）；College of Computing and Data Science, Nanyang Technological University, Singapore（计算与数据科学学院，南洋理工大学，新加坡）
Forget-MI: Machine Unlearning for Forgetting Multimodal Information in Healthcare Settings
Department of Machine Learning, Mohamed bin Zayed University of Artificial Intelligence, Abu Dhabi, UAE Department of Computer Vision, Mohamed bin Zayed University of Artificial Intelligence, Abu Dhabi, UAE Department of Computer Science, Michigan State University, United States 0.2cm
Class-wise Federated Unlearning: Harnessing Active Forgetting with Teacher-Student Memory Generation
School of Communication Engineering, Hangzhou Dianzi University（杭州电子科技大学通信工程学院）；College of Computer Science & Technology, Zhejiang University（浙江大学计算机科学与技术学院）；School of Cyberspace, Hangzhou Dianzi University（杭州电子科技大学网络空间学院）
HKUST（香港科技大学）；University of Cambridge（剑桥大学）；Central South University（中南大学）；Shanghai Jiaotong University（上海交通大学）；SenseTime Research（商汤科技研究院）；Tsinghua University（清华大学）
FaLW: A Forgetting-aware Loss Reweighting for Long-tailed Unlearning
University of Science and Technology of China(USTC)（中国科学技术大学）；Suzhou Institute for Advanced Research, USTC（中国科学技术大学苏州研究院）；City University of Hong Kong（香港城市大学）；Tongji University（同济大学）
Decoupling the Class Label and the Target Concept in Machine Unlearning
Hong Kong Baptist University（香港浸会大学）；RIKEN Center for Advanced Intelligence Project（理化学研究所先进智能中心）；Shanghai Jiao Tong University（上海交通大学）；Shanghai AI Laboratory（上海人工智能实验室）；The University of Tokyo（东京大学）
Cornell University（康奈尔大学）；University of Waterloo（滑铁卢大学）；Google Research（谷歌研究院）
We study the problem of unlearning datapoints from a learnt model. The learner first receives a dataset $S$ drawn i.i.d. from an unknown distribution, and outputs a model $\widehat{w}$ that performs well on unseen samples from the same distribution. However, at some point in the future, any training datapoint $z \in S$ can request to be unlearned, thus prompting the learner to modify its output model while still ensuring the same accuracy guarantees. We initiate a rigorous study of generalization in machine unlearning, where the goal is to perform well on previously unseen datapoints. Our focu…
Is Gradient Ascent Really Necessary? Memorize to Forget for Machine Unlearning
Sydney AI Centre, The University of Sydney（悉尼人工智能中心，悉尼大学）；Hong Kong Baptist University（香港 Baptist 大学）；RIKEN AIP（日本理化学研究所AIP）；University of Technology Sydney（悉尼技术大学）
Nowadays, machine learning models, especially neural networks, become prevalent in many real-world applications.These models are trained based on a one-way trip from user data: as long as users contribute their data, there is no way to withdraw; and it is well-known that a neural network memorizes its training data. This contradicts the "right to be forgotten" clause of GDPR, potentially leading to law violations. To this end, machine unlearning becomes a popular research topic, which allows users to eliminate memorization of their private data from a trained machine learning model.In this pap…
Electrical Engineering and Computer Science, University of California, Merced, CA 95343, USA（电气工程与计算机科学系，加州大学默塞德分校）；Department of Applied Mathematics, University of California, Merced, CA 95343, USA（应用数学系，加州大学默塞德分校）；Department of Innovation and Research, North Carolina State University, Raleigh, NC, USA（创新与研究系，北卡罗来纳州立大学）；Automation (MESA) Lab, Department of Mechanical Engineering, School of Engineering, University of California, Merced, CA 95343, USA（自动化（MESA）实验室，机械工程系，工程学院，加州大学默塞德分校）
