驯服长尾：通过自适应扰动重平衡对抗训练
Taming the Long Tail: Rebalancing Adversarial Training via Adaptive Perturbation
把对抗扰动当作可调节的训练分布重加权手段，为长尾视觉任务（如长尾分类、检测）提供了一种即插即用的对抗训练改造思路，可直接嵌入现有对抗训练流程。
原文依据 we show that perturbations can simultaneously address both adversarial vulnerability and class imbalance
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论深度神经网络极易受到对抗样本的影响，即能够显著降低模型性能的小扰动。尽管对抗训练已成为主要的防御策略，但多数研究聚焦于平衡数据集，忽视了真实世界长尾数据带来的挑战。受对抗样本中的扰动本质上会改变训练分布这一事实的启发，我们从理论上研究了其影响。我们首先重新审视了面向长尾数据的对抗训练，并识别出两个关键局限：(i) 类别不平衡导致的训练目标偏斜，以及 (ii) 对抗分布演化不稳定。此外，我们表明扰动能够同时缓解对抗脆弱性与类别不平衡。基于这些洞见，我们提出了 METHODNAME，一个在对抗训练过程中自适应调整扰动的即插即用框架。大量实验表明，METHODNAME 在长尾数据集上持续提升了对抗鲁棒性与类别平衡。代码见 https://github.com/zhang-lilin/RobustLT 。
英文原文摘要 1188 字符
Deep neural networks are highly vulnerable to adversarial examples, i.e.,small perturbations that can significantly degrade model performance. While adversarial training has become the primary defense strategy, most studies focus on balanced datasets, overlooking the challenges posed by real-world long-tail data. Motivated by the fact that perturbations in adversarial examples inherently alter the training distribution, we theoretically investigate their impact. We first revisit adversarial training for long-tail data and identify two key limitations: (i) a skewed training objective caused by class imbalance, and (ii) unstable evolution of adversarial distributions. Furthermore, we show that perturbations can simultaneously address both adversarial vulnerability and class imbalance. Based on these insights, we propose \METHODNAME , a plug-and-play framework that adaptively adjusts perturbations during adversarial training. Extensive experiments demonstrate that \METHODNAME consistently enhances adversarial robustness and class-balance on long-tailed datasets. The code is available at \href https://github.com/zhang-lilin/RobustLT https://github.com/zhang-lilin/RobustLT .