arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2606.03381cs.CRcs.AI
收藏

AI模型提取攻击:绕过防御中的单客户端假设

AI Model Extraction Attacks: Bypassing Single-Client Assumptions in Defenses

ETH Zurich(苏黎世联邦理工学院);University of Zurich(苏黎世大学);University of Tübingen(图宾根大学)

Maxime Schwarzer, Johannes F. Loevenich, Gustavo Sánchez, Laurin Holz, Thies Möhlenhof, Tobias Hürten, Roberto Rigolin F. Lopes, Veit Hagenmeyer

AI总结:

本文通过提出CerberusAI框架,系统性地证明模型提取攻击中的单客户端假设(SCA)在高级持续性威胁(APT)等协同攻击者面前无效,并展示基本轮询查询分布策略即可绕过PRADA等防御机制,呼吁转向无状态、独立于身份的防御架构。

摘要:

确保部署在军事指挥控制(C2)系统和关键基础设施中的人工智能(AI)模型的保护对于维持信息优势至关重要。模型提取攻击(MEA)构成了重大威胁,因为它们使对手能够复制专有模型、泄露受保护信息并准备离线对抗性攻击。然而,当前的防御策略主要依赖于单客户端假设(SCA),即隐含地假设攻击源自孤立身份。本工作系统地证明了在协同威胁行为者(如高级持续性威胁APT)存在的情况下,SCA从根本上无效。我们引入了一个模块化、开源框架CerberusAI,用于可复现的模型窃取研究,并利用它模拟分布式攻击场景。我们的实证评估表明,成熟的防御机制(如防止深度神经网络模型窃取攻击PRADA)可以通过基本的轮询查询分布策略被绕过,导致检测性能显著下降。此外,我们证明即使是全局聚合方法也可以通过自适应流量混合使其在操作上变得无用。这些结果强调了在模型提取攻击领域需要向有状态、独立于身份的防御架构进行范式转变。本文最初发表于由信息系统技术(IST)科学与技术委员会IST-224-RSY组织的国际军事通信与信息系统会议(ICMCIS),该会议于2026年5月12-13日在英国巴斯举行,并获得了最佳论文奖。

英文摘要:

Ensuring the protection of Artificial Intelligence (AI) models deployed in military Command and Control (C2) systems and critical infrastructure is essential for maintaining information superiority. Model Extraction Attacks (MEAs) pose a significant threat, as they enable adversaries to replicate proprietary models, compromise protected information, and prepare offline adversarial attacks. However, current defense strategies predominantly rely on the Single Client Assumption (SCA), which is the implicit assumption that attacks originate from isolated identities. This work systematically demonstrates that the SCA is fundamentally invalid in the presence of coordinated threat actors, such as Advanced Persistent Threats (APTs). We introduce a modular, open-source framework called CerberusAI for reproducible model-stealing research, and use it to simulate distributed attack scenarios. Our empirical evaluation shows that well-established defense mechanisms, such as Protecting Against Deep Neural Network Model Stealing Attacks (PRADA), can be bypassed by basic round-robin query distribution strategies, resulting in a significant reduction in detection performance. Furthermore, we demonstrate that even global aggregation approaches can be rendered operationally useless through adaptive traffic mixing. These results highlight the need for a paradigm shift towards stateful, identity-independent defense architectures in the field of model extraction attacks. This paper was originally presented at the International Conference on Military Communication and Information Systems (ICMCIS), organized by the Information Systems Technology (IST) Scientific and Technical Committee, IST-224-RSY - the ICMCIS, held in Bath, United Kingdom, 12-13 May 2026 and won the best paper award.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑