English(EN) BadMoE: Backdooring Mixture-of-Experts LLMs via Routing-Aware Triggers and Task-Underutilized Experts

新的BadMoE攻击通过路由操纵针对混合专家LLM

作者 PulseAugur 编辑部 · [1 个来源] · 2026-11-11 05:00

研究人员开发了一种名为BadMoE的新方法来利用混合专家（MoE）大型语言模型的漏洞。该技术针对MoE架构的路由机制，该机制选择性地激活专家子集来处理输入。BadMoE识别利用不足的专家，并使用精心设计的触发器来操纵它们的激活，从而将预定义的后门行为注入模型。该攻击已在Mixtral、OLMoE和Deepseek等模型上针对分类和定向拒绝等各种任务进行了演示。 AI

影响突出了MoE模型中一类新的漏洞，可能影响其在部署中的安全性和可信度。

排序理由详细介绍LLM架构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: 34 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 详细介绍LLM架构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) · Qingyue Wang, Qi Pang, Xixun Lin, Shuai Wang, Daoyuan Wu · 2026-11-11 05:00

BadMoE：通过路由感知触发器和任务利用不足的专家后门攻击混合专家大语言模型

arXiv:2504.18598v3 Announce Type: replace-cross Abstract: Mixture-of-Experts (MoE) architectures are rapidly becoming the standard for building scalable and efficient large language models (LLMs), thanks to their routing mechanism that activates only a subset of experts for each …

本文如何被排名

报道来源 [1]

BadMoE：通过路由感知触发器和任务利用不足的专家后门攻击混合专家大语言模型

相关实体

相关话题