面向长尾学习的决策边界感知生成
Decision Boundary-aware Generation for Long-tailed Learning
把扩散生成增强从「补数量」推进到「补位置」：通过生成靠近决策边界的样本缓解头尾迁移带来的潜变量非局部特征混合，可直接复用到任何长尾分类或检测的生成式数据增强流程中。
原文依据 we first identify the problem of boundary ambiguity and then propose Decision Boundary-aware Generation (DBG) framework
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论长尾数据会使决策边界偏向头部类别，并降低尾部类别的准确率。基于扩散的生成式增强通过生成额外数据来缓解这一问题，而头到尾迁移进一步减轻了生成器从长尾数据集中继承的偏差。然而，我们表明，头到尾迁移虽然有助于平衡分类器的决策空间，却也会引入潜变量中的非局部特征混合，使类间特征相互纠缠，从而导致决策边界重叠与尾部类别分布偏移。为解决这一问题，我们首先识别出边界模糊这一问题，随后提出决策边界感知生成（DBG）框架，通过生成信息量丰富的近边界样本来促进近边界表征学习。总体而言，DBG 在重新平衡长尾数据集的同时，为长尾学习带来更可分的决策空间。在标准长尾基准上，DBG 在减少类间重叠的同时，持续提升尾部类别与整体准确率。DBG 的代码已公开于 https://github.com/keepdigitalabc-svg/DBG。
英文原文摘要 1103 字符
Long-tailed data bias decision boundaries toward head classes and degrade tail class accuracy. Diffusion-based generative augmentation address this problem by generating additional data, while head-to-tail transfer further mitigate the generator bias inherit from long-tailed dataset. However, we show that while head-to-tail transfer helps balance the decision space of the classifier, it also induces latent non-local feature mixing that entangles inter-class features, causing decision boundary overlap and tail class distribution shift. To address this, we first identify the problem of boundary ambiguity and then propose Decision Boundary-aware Generation (DBG) framework, which promotes near-boundary representation learning by generating informative near-boundary samples. Overall, DBG rebalances the long-tailed dataset while yielding more separable decision space for long-tailed learning. Across standard long-tailed benchmarks, DBG consistently improves tail class and overall accuracy with less inter-class overlap. The code of DBG is available at https://github.com/keepdigitalabc-svg/DBG.