针对跨架构攻击效果差的问题，提出频谱感知的梯度优化方法提升对抗样本迁移能力
Season: Spectrum-Aware Orthogonal Gradient Refinement for Transfer-Based Adversarial Attacks
- 将梯度分解为结构与纹理两个频段，分别处理以适配CNN和ViT差异
- 在统一测试下平均提升6.6个百分点的攻击成功率，最高达16.0点
- 无需训练，可直接增强现有攻击方法，适合黑盒攻击研究者
跨架构的迁移式对抗攻击常因卷积神经网络偏好局部纹理而视觉变换器依赖全局形状而导致效果不佳。本文提出Season，一种面向L-infinity迁移攻击的频谱感知正交梯度优化框架，基于白盒代理模型对ImageNet上的黑盒目标模型进行攻击。Season将每次更新分解为低频（结构线索）与高频（纹理）分支。通过低显著性引导策略，将高频能量重新分配至背景区域，保留视觉变换器所依赖的前景结构。再通过正交投影，强制纹理更新位于结构方向的正交补空间，减少特征干扰。作为无训练、即插即用的封装模块，Season可增强八种梯度稳定与输入增强类攻击，且不修改其核心逻辑。在八个不同类型的模型（包括CNN、ViT和MLP）上，Season在统一协议下平均提升6.6个百分点的攻击成功率，最高提升达16.0个百分点。
原文摘要 · Abstract (English)
Transfer-based adversarial attacks often transfer poorly across heterogeneous architectures because CNNs favor local textures while Vision Transformers (ViTs) rely on global shapes. We propose Season, a spectrum-aware orthogonal gradient refinement framework for L-infinity transfer attacks against black-box target models on ImageNet, using a white-box surrogate. Season decomposes each update into a low-frequency branch capturing structural cues and a high-frequency branch capturing textures. A low-saliency guidance scheme reallocates high-frequency energy to background regions, preserving foreground structures that ViTs depend on. An orthogonal projection then forces the textural update to lie in the orthogonal complement of the structural direction, mitigating feature interference. As a training-free plug-and-play wrapper, Season enhances eight gradient-stabilization and input-enhancement attacks without modifying their cores. Across eight CNN, ViT, and MLP targets, Season improves transfer success rate by 6.6 percentage points on average and up to 16.0 points over strong baselines under a unified protocol.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。