arXiv cs.CL· Guangjing Wang, Ce Zhou, Yuanda Wang, Bocheng Chen, Hanqing Guo, Qiben Yan·· 12 小时前AI 评分31
跨越边界的可迁移攻击：AI 系统可迁移攻击综述
Beyond Boundaries: A Comprehensive Survey of Transferable Attacks on AI Systems
AI 导读
一篇综述提出覆盖跨实例、跨域、跨模态、跨模型、跨任务、跨部署六个维度的统一分类法，用于比较 evasion、投毒、后门、隐私推断、模型窃取与基于提示词的攻击之间的可迁移性。文章将梯度对齐、共享表征、训练持久性与部署不变性等迁移机制，与基于增强、解耦和优化的增强方法相连接，并指出迁移效果不稳定、指标不兼容以及模拟与物理评估之间的差距。
正文 · 原文
Abstract:Transferable attacks allow adversarial manipulations or learned attack mechanisms constructed in one setting to compromise other AI systems, potentially reducing attack costs and amplifying security and privacy risks. Yet fragmented terminology and evaluation practices hinder understanding of what transfers, why it transfers, and where transfer fails. This survey synthesizes representative research and develops a unified taxonomy spanning six complementary axes: cross-instance, cross-domain, cross-modality, cross-model, cross-task, and cross-deployment. By separating source-target configuration changes from attacker access and adaptation budgets, the taxonomy provides a common framework for comparing transfer across evasion, poisoning, backdoors, privacy inference, model stealing, and prompt-based attacks. We connect transfer mechanisms, including gradient alignment, shared representations, training persistence, and deployment invariances, to enhancement methods based on augmentation, disentanglement, and optimization. Our synthesis identifies inconsistent transfer, incompatible metrics, and gaps between simulated and physical evaluations, and establishes a research agenda for compositional transfer, configuration differences, temporal persistence, and reproducible benchmarking. Together, this work supports systematic investigation of shared vulnerabilities and guides transfer-aware security assessment and defense.
| Subjects: | Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2311.11796 [cs.CR] |
| (or arXiv:2311.11796v3 [cs.CR] for this version) | |
| https://doi.org/10.48550/arXiv.2311.11796 arXiv-issued DOI via DataCite |
Submission history
From: Guangjing Wang [view email]
[v1]
Mon, 20 Nov 2023 14:29:45 UTC (154 KB)
[v2]
Mon, 12 May 2025 11:25:11 UTC (260 KB)
[v3]
Wed, 14 Oct 2026 18:38:03 UTC (172 KB)
来源：arXiv cs.CL · arxiv.org