False Floors: LLM Safety Routing Evaluations Break Under Distribution Shift
这篇论文研究的是 LLM 安全路由(safety routing)评测中的方法论缺陷。所谓安全路由,是指用一个路由器把每个用户请求分派给池中若干模型之一,通常以「池中最好的单一模型」作为对照基线来判断路由器是否真的带来了收益。作者指出,主流
今日共收录 931 条安全情报,包含 128 个 CVE,457 篇安全通告,以及 95 篇研究论文。
← 返回归档列表这篇论文研究的是 LLM 安全路由(safety routing)评测中的方法论缺陷。所谓安全路由,是指用一个路由器把每个用户请求分派给池中若干模型之一,通常以「池中最好的单一模型」作为对照基线来判断路由器是否真的带来了收益。作者指出,主流
该论文来自 arXiv(分类 cs.CL、cs.AI),作者为 Pengfei Li、Naufal Suryanto、Sicheng Zhang、Muzammal Naseer。研究背景是:LLM 正被大量引入网络安全工作流,用来把分析员的
本论文关注大语言模型(LLM)安全对齐中的机器遗忘(machine unlearning)机制。移除模型中的危险知识、实现安全对齐通常依赖遗忘技术,但近期研究揭示了一个持续存在的安全风险:被「遗忘」的模型在面对再训练攻击(retrainin
该论文是一篇面向区块链辅助检测与响应体系的知识系统化(SoK)综述。研究背景是:区块链辅助的入侵检测与防御系统(IDS/IPS)在物联网(IoT)与工业物联网(IIoT)场景下的文献已相当成熟,但既有系统性综述存在两个关键缺陷——第一,忽视
Cloud & Application Security
From Scanner to Stealer: Inside the trivy-action Supply Chain Compromise
Next-Gen Identity Security
CrowdStrike Expands Identity Leadership with OpenID and IDPro
cloud architecture assessment
Google Cloud Next 2026Catch-up sessions on the keynotes and select sessions are now available on demand.Explore content
The Identity Problem Hiding in AI Agent Deployments
94% of Organizations Report Cloud Breaches: CrowdStrike State of CDR Survey
Falcon Cloud Security June 2026 Release: Updates for Azure and Google Cloud
Falcon Cloud Security July 2026 Release: Helping Security Teams Move Faster in the Cloud
CrowdStrike Named Strongest Overall Leader in 2026 Frost Radar™: Cloud Workload Protection Platforms
CrowdStrike Extends Endpoint Security to Stop Software Supply Chain Attacks
CrowdStrike Announces Agentic Identity Provider
ransomware and multifaceted extortion defense
The Technical University of Denmark (DTU) says information belonging to up to 200,000 users may have been exposed after
Featuring: Cybersecurity is being reshaped by the expansion of cloud infrastructure, AI, distributed systems, and incre