在核心处擦除：用于机器擦除的表示性遗忘
Erase at the Core: Representation Unlearning for Machine Unlearning
AI总结：
Erase at the Core通过多层对比性擦除和深度监督学习，有效减少模型表示相似性，提升机器擦除的表示层面遗忘能力。
摘要：
许多近似机器擦除方法在logit层面表现出强烈的遗忘效果——例如在遗忘集上接近零的准确性——但仍然在内部特征表示中保留了大量信息。我们将这种差异称为表面遗忘。最近的研究表明，现有大多数擦除方法主要改变最终分类器，使中间表示几乎未变，并且高度相似于原始模型。为了解决这一限制，我们引入了Erase at the Core（EC），一种旨在在整个网络层次结构中强制遗忘的框架。EC通过在遗忘集上进行多层对比性擦除，并通过深度监督学习对保留集进行保护。具体而言，EC在中间层附加辅助模块，并在每个监督点应用对比性擦除和交叉熵损失，采用层间加权损失。实验结果表明，EC不仅实现了有效的logit层面遗忘，还显著减少了中间层与原始模型之间的表示相似性。此外，EC是模型无关的，可以作为插件模块整合到现有擦除方法中，从而在保持保留集性能的同时提高表示层面的遗忘效果。
英文摘要：
Many approximate machine unlearning methods demonstrate strong logit-level forgetting -- such as near-zero accuracy on the forget set -- yet continue to preserve substantial information within their internal feature representations. We refer to this discrepancy as superficial forgetting. Recent studies indicate that most existing unlearning approaches primarily alter the final classifier, leaving intermediate representations largely unchanged and highly similar to those of the original model. To address this limitation, we introduce the Erase at the Core (EC), a framework designed to enforce forgetting throughout the entire network hierarchy. EC integrates multi-layer contrastive unlearning on the forget set with retain set preservation through deeply supervised learning. Concretely, EC attaches auxiliary modules to intermediate layers and applies both contrastive unlearning and cross-entropy losses at each supervision point, with layer-wise weighted losses. Experimental results show that EC not only achieves effective logit-level forgetting, but also substantially reduces representational similarity to the original model across intermediate layers. Furthermore, EC is model-agnostic and can be incorporated as a plug-in module into existing unlearning methods, improving representation-level forgetting while maintaining performance on the retain set.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
