English(EN) Inducing Overthink: Hierarchical Genetic Algorithm-based DoS Attack on Black-Box Large Language Reasoning Models

新型攻击诱导大型语言模型推理模型“过度思考”

作者 PulseAugur 编辑部 · [1 个来源] · 2026-05-13 10:57

研究人员开发了一种新技术，利用大型语言推理模型（LRMs）中的一个漏洞，使其“过度思考”。该技术使用分层遗传算法生成输入，导致推理过程过长且冗余，从而增加了延迟和资源消耗。该攻击在MATH基准测试中显示输出长度显著增加，最高可达26.1倍，并对各种最先进的模型显示出有效性，凸显了改进针对此类拒绝服务攻击的防御措施的必要性。 AI

影响这项研究揭示了大型语言模型推理中的一个新漏洞，可能影响依赖它们的AI系统的可靠性和可用性。

排序理由该集群包含一篇详细介绍针对大型语言模型的新型攻击方法的学术论文。[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 →

本文如何被排名

Signal score: 0 / 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针对大型语言模型的新型攻击方法的学术论文。[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: 111 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.AI TIER_1 English(EN) · Zhixuan Chu · 2026-05-13 10:57

诱导过度思考：基于分层遗传算法的黑盒大型语言推理模型DoS攻击

Large Reasoning Models (LRMs) are increasingly integrated into systems requiring reliable multi-step inference, yet this growing dependence exposes new vulnerabilities related to computational availability. In particular, LRMs exhibit a tendency to "overthink", producing excessiv…

本文如何被排名

报道来源 [1]

诱导过度思考：基于分层遗传算法的黑盒大型语言推理模型DoS攻击

相关实体

相关话题