arXivDaily arXiv每日学术速递 周一至周五更新

AI 大模型

大模型对齐与安全

大模型对齐、安全、越狱、红队、提示注入和可信评测。

2026-04-23 至 2026-04-23 共收录 2 信号源:cs.CL, cs.AI, cs.CY, cs.LG

1. 越狱攻击 2 篇

2604.20771 2026-04-23 cs.CR cs.AI 57%

DAIRE: A lightweight AI model for real-time detection of Controller Area Network attacks in the Internet of Vehicles

DAIRE:一种轻量级AI模型,用于实时检测车联网中的控制器局域网络攻击

Shahid Alam, Amina Jameel, Zahida Parveen, Ehab Alnfrawy, Adeela Ashraf, Raza Uddin, Jamal Aqib

机构 * Department of Information Security, College of Computer Science and Engineering, University of Ha’il(信息安全系,计算机科学与工程学院,哈伊尔大学) Department of Computer Engineering, Center of Excellence in Artificial Intelligence, Bahria University(计算机工程系,人工智能卓越中心,巴赫里大学) Department of Artificial Intelligence, College of Computer Science and Engineering, University of Ha’il(人工智能系,计算机科学与工程学院,哈伊尔大学) Department of Computer Science, College of Computer Science and Engineering, University of Ha’il(计算机科学系,计算机科学与工程学院,哈伊尔大学) Department of Software Engineering, College of Computer Science and Engineering, University of Ha’il(软件工程系,计算机科学与工程学院,哈伊尔大学)

专题命中 越狱攻击 :safety(abstract);分类 cs.AI

AI总结 本文提出DAIRE模型,利用轻量级神经网络实现实时检测和分类CAN攻击,实验显示其在准确率和速度上均优于现有方法。

Journal ref Machine Learning with Applications (2026): 100859

详情

展开后加载摘要…

URL PDF HTML 收藏
2404.07527 2026-04-23 cs.CR 50%

Security Modelling for Cyber-Physical Systems: A Systematic Literature Review

面向网络物理系统的安全建模:系统文献综述

Shaofei Huang, Christopher M. Poskitt, Lwin Khin Shar

专题命中 越狱攻击 :safety(abstract)

AI总结 本文系统回顾了网络物理系统安全建模的研究,探讨威胁与攻击建模的区别及其对系统安全的影响,总结现有研究的局限性。

Comments Accepted by ACM Transactions on Cyber-Physical Systems (TCPS)

Journal ref ACM Trans. Cyber Phys. Syst. 10(2), Article 20, 1-29 (2026)

详情

展开后加载摘要…

URL PDF HTML 收藏