English(EN) Forget the Data, Keep the Capability: Switchable Unlearning for Retrieval

新的MoME框架实现了AI隐私的可切换遗忘

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-28 04:00

研究人员开发了MoME，一个旨在实现重新识别模型隐私合规的新型框架。这个可切换系统允许单个预训练骨干模型在两种模式下运行：‘完整’模式用于标准推理，‘核心’模式用于遗忘式推理。MoME通过分离保留检索能力的核心参数和存储完整数据知识的寄生参数来实现这一点，确保核心参数不受数据删除请求的影响。这种方法允许选择性地遗忘个体，同时保持模型识别他人的能力，其性能已在Market-1501、VeRi-776和RegDB等数据集上得到验证。 AI

影响该框架通过允许选择性数据删除而无需完全重新训练，可以增强AI系统的隐私合规性。

排序理由该集群包含一篇详细介绍AI模型遗忘新框架的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CV 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 11 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍AI模型遗忘新框架的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, safety; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Same-day; Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CV TIER_1 English(EN) · Jingong Chen, Kyle Gao, Guangyuan Li, Chulwoo Pack, Sijia Liu, Kwanghee Won · 2026-10-28 04:00

忘掉数据，保留能力：可切换的检索式非学习

arXiv:2610.36858v1 Announce Type: new Abstract: Privacy compliance requires re-identification models to selectively forget individuals while preserving their ability to recognize others. For this, we present MoME, a switchable framework that can adapt pre-trained backbones for pr…

本文如何被排名

报道来源 [1]

忘掉数据，保留能力：可切换的检索式非学习

相关实体

相关话题