论文概要
研究领域: ML 作者: Leonardo Santiago Benitez Pereira, Marcos Escudero Viñolo, Luis Herranz Arribas 发布时间: 2026-09-03 arXiv: 2509.00002
中文摘要
机器遗忘研究如何从AI模型中移除知识，使系统忘记它之前学习的概念。尽管生成式机器遗忘进展迅速，但对本应保留的语义相关概念的意外退化（即干扰）仍然缺乏充分表征和一致评估。本文介绍了I-CARE，一种将干扰形式化为生成式遗忘中一级研究对象的方法论。I-CARE并非提出新的基准或遗忘算法，而是为任务、指标和结果报告模板提供形式化定义，从而实现跨遗忘设置对干扰的系统性和可重复研究。虽然我们的方法论旨在随着模型和遗忘算法的演进而保持有效，将长期科学洞察与瞬时的经验结果解耦，但我们使用最先进的算法和常用数据集进行了可行性演示。结果表明，I-CARE能够跨多种遗忘设置对干扰模式进行有意义的分析，确立了该框架的实际适用性。该方法的软件实现以开源框架形式提供，并附带基于Web的图形界面，无需直接与代码库交互或使用专门的数据分析工具即可探索本研究的结果。
原文摘要
Machine unlearning studies the removal of knowledge from an AI model, making the system forget a concept it previously learned. Despite rapid progress in generative machine unlearning, the unintended degradation of semantically related concepts that should have been retained (henceforth, interference) remains poorly characterized and inconsistently evaluated. This paper introduces I-CARE, a methodology that formalizes interference as a first-class object of study in generative unlearning. Rather than proposing a new benchmark or unlearning algorithm, I-CARE provides formal definitions for tasks, metrics, and templates for reporting results, enabling the systematic and reproducible study of interference across unlearning settings. While our methodology is designed to remain valid as models …
— 自动采集于 2026-09-03
#论文 #arXiv #ML #小凯