PulseAugur
实时 09:31:28
English(EN) Alignment Whack-a-Mole : Finetuning Activates Verbatim Recall of Copyrighted Books in Large Language Models

研究发现大型语言模型可被微调以逐字回忆受版权保护的书籍

一项新的研究论文揭示,微调大型语言模型可能会无意中导致它们逐字回忆受版权保护的材料,尽管人工智能公司保证他们的模型不存储训练数据。研究人员演示了通过训练模型扩展情节摘要,像GPT-4o、Gemini 2.5 Pro和DeepSeek-V3.1这样的模型可以重现高达90%的受版权保护的书籍。这种漏洞似乎是行业普遍存在的,因为来自不同提供商的不同模型在相同数据区域表现出类似的记忆模式。 AI

影响 揭示了大型语言模型中一种潜在的行业普遍漏洞,可能破坏版权侵权诉讼的法律辩护。

排序理由 发表在arXiv上的研究论文,详细介绍了大型语言模型的一个新漏洞。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.AI 阅读 →

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

研究发现大型语言模型可被微调以逐字回忆受版权保护的书籍

本文如何被排名

Signal score
13 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
发表在arXiv上的研究论文,详细介绍了大型语言模型的一个新漏洞。[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
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准

报道来源 [1]

  1. arXiv cs.AI TIER_1 English(EN) · Xinyue Liu, Niloofar Mireshghallah, Jane C. Ginsburg, Tuhin Chakrabarty ·

    对齐打地鼠:微调激活大型语言模型对受版权书籍的逐字回忆

    arXiv:2603.20957v4 Announce Type: replace-cross Abstract: Frontier LLM companies have repeatedly assured courts and regulators that their models do not store copies of training data. They further rely on safety alignment strategies via RLHF, system prompts, and output filters to …