English(EN) Rethinking Latency Denial-of-Service: Attacking the LLM Serving Framework, Not the Model

新攻击目标大语言模型服务框架，绕过模型防御

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-09 04:00

研究人员开发了一种新的方法来攻击大语言模型（LLM）的服务框架，而不是模型本身。这种“填充与挤压”（Fill and Squeeze）策略通过耗尽 KV 缓存并强制重复抢占来攻击调度器的状态转换。该攻击在 vLLM 框架上显著降低了首次令牌生成时间（TTFT）和每个输出令牌的时间，在实际的黑盒环境中以比以前更低的成本证明了其有效性。 AI

影响这项研究突显了大语言模型服务基础设施中一类新的漏洞，可能影响部署安全性和成本效益。

排序理由详细介绍大语言模型服务框架新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: 18 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 详细介绍大语言模型服务框架新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: 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]

arXiv cs.AI TIER_1 English(EN) · Tianyi Wang, Huawei Fan, Yuanchao Shu, Peng Cheng, Cong Wang · 2026-10-09 04:00

重新思考延迟拒绝服务攻击：攻击大型语言模型服务框架而非模型本身

arXiv:2602.07878v2 Announce Type: replace-cross Abstract: LLM inference is inherently expensive, even a modest slowdown can translate into substantial operating costs and severe availability risks. Recently, a growing body of research known as latency attacks focuses on crafting …

本文如何被排名

报道来源 [1]

重新思考延迟拒绝服务攻击：攻击大型语言模型服务框架而非模型本身

相关实体

相关话题