English(EN) Taming the Long Tail: Rebalancing Adversarial Training via Adaptive Perturbation

新框架 RobustLT 解决了不平衡数据集上的对抗性训练问题

作者 PulseAugur 编辑部 · [1 个来源] · 2026-05-13 11:51

研究人员开发了一个名为 RobustLT 的新框架，以改进深度神经网络的对抗性训练，特别是在具有长尾分布的数据集上。该框架解决了当前方法中的局限性，例如由于类别不平衡导致的训练目标倾斜和不稳定的对抗性分布。通过在训练过程中自适应地调整扰动，RobustLT 旨在增强对抗性鲁棒性和类别平衡性，这已在大量实验中得到证明。 AI

影响提高了模型在不平衡数据集上的鲁棒性，可能提高其在实际应用中的可靠性。

排序理由介绍对抗性训练新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CV 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 介绍对抗性训练新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, safety; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: 105 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CV TIER_1 English(EN) · Xianggen Liu · 2026-05-13 11:51

驯服长尾：通过自适应扰动重新平衡对抗性训练

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 p…

本文如何被排名

报道来源 [1]

驯服长尾：通过自适应扰动重新平衡对抗性训练

相关实体

相关话题