English(EN) Intent-Hiding Jailbreaks: An Information-Theoretic Framework for Compositional Attacks

新框架使用信息论分析大语言模型越狱

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-05 04:00

研究人员开发了一个信息论框架，用于分析和理解大语言模型中的“隐藏意图越狱”。这些攻击通过将有害请求嵌入到更大、看似良性的查询中来起作用，从而使模型更可能遵从。该框架通过将任务与被判断为有害的概率相关联来建模，目标是在包含有害目标时也能维持估计的意图。该研究探讨了查询无关和查询相关的设置，发现组合查询确实可以在一定的搜索预算内引发超出直接请求的目标行为，尽管在某些模型中较大的捆绑包大小会降低目标保留率。 AI

影响这项研究为理解和潜在缓解针对大语言模型的复杂越狱攻击提供了一个理论框架。

排序理由关于大语言模型安全研究的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 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: 6 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 关于大语言模型安全研究的学术论文。[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, 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.LG TIER_1 English(EN) · Fengwei Tian, Ravi Tandon · 2026-10-05 04:00

隐藏意图的越狱：一种用于组合攻击的信息论框架

arXiv:2610.02302v1 Announce Type: cross Abstract: Recent work has shown that large language models (LLMs) can be vulnerable to jailbreak attacks in which harmful intent is obscured through composition with benign tasks. A harmful request refused in isolation may elicit a differen…

本文如何被排名

报道来源 [1]

隐藏意图的越狱：一种用于组合攻击的信息论框架

相关实体

相关话题