English(EN) When Forgetting is not Catastrophic: On the Mechanics of Spurious Forgetting

新研究详述语言模型的虚假遗忘现象

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-07 04:00

研究人员在语言模型中发现了一种称为虚假遗忘的现象，即在微调过程中知识似乎丢失，但实际上可以恢复。这种遗忘甚至可以自我逆转，旧的事实最初会消失，然后重新出现，最后永久消失。一个包含键的共享结构、集中的新值和网络归一化的最小联想记忆模型重现了这些动态。研究表明，遗忘涉及可逆的、共享的访问丢失以及更慢的、灾难性的个体事实侵蚀，两者的主导地位取决于新数据如何影响旧记忆。 AI

影响这项研究揭示了语言模型遗忘的内部机制，可能为未来的微调策略和模型可解释性提供信息。

排序理由该集群包含一篇详细介绍语言模型机制新发现的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 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语言模型机制新发现的学术论文。[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, model release; 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.CL TIER_1 English(EN) · Vedant Palit, Florent Draye, Nicolas Zucchet, Zhijing Jin, Bernhard Sch\"olkopf · 2026-10-07 04:00

遗忘并非灾难：关于虚假遗忘的机制

arXiv:2610.08718v1 Announce Type: new Abstract: Knowledge that a language model appears to forget during finetuning often remains stored and can be recovered, a phenomenon called spurious forgetting. Finetuning on new facts can even produce forgetting that undoes itself: recall o…

本文如何被排名

报道来源 [1]

遗忘并非灾难：关于虚假遗忘的机制

相关实体

相关话题