你的差异性定义了你：利用类别多样性进行互补学习
Your Dissimilarities Define You: Complementary Learning Exploiting Class Diversities
把「非目标类」整体建模为一个 opposite-class 并对其激活分布做显式约束，可作为分类头/损失层的即插即用正则项，配合神经坍缩分析改善特征几何，适用于开放集、少样本与域泛化的视觉分类骨干训练。
原文依据 we introduce the concept of an opposite-class, which consists of everything that is not part of a corresponding class
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论本文利用类别差异性，提供超越正确分类之外的互补学习信息，而现有学习范式并未充分利用这一信息。为建模这些差异性，我们引入了「对立类」的概念，它由所有不属于对应类别的部分构成，即来自非目标类的全部样本或来自未知类的样本。通过在非目标类上设置适当编码的目标分布，我们显式优化模型在所有非目标类上的激活分布，从而增强类别差异性信息，并能够更好地控制表征的几何结构。我们从理论和实验两方面分析了所提方法的收敛动态，表明它自然地将表征推向神经坍缩，从而得到更具判别性且更鲁棒的特征。我们在多种分类设定下的广泛评估表明，该方法在闭集分类、开集分类、少样本分类和域泛化上均取得一致的提升。代码：https://github.com/katdimitris/Complementary-Dissimilarity-Loss
英文原文摘要 1189 字符
In this work, we exploit class dissimilarities to provide complementary learning information beyond correct classification, that is not fully utilized in existing learning paradigms. To model these dissimilarities, we introduce the concept of an opposite-class, which consists of everything that is not part of a corresponding class, i.e., all samples from non-target classes or samples from unknown classes. By setting appropriately encoded target distributions over the non-target classes, we explicitly optimize the model's activation distributions across all non-target classes, which enhances class dissimilarity information and enables better control over the geometry of the representations. We analyze the convergence dynamics of our proposed approach, both theoretically and empirically, showing that it naturally pushes the representations towards neural collapse, leading to more discriminative and robust features. Our extensive evaluation across multiple classification settings demonstrates consistent improvements of our method on closed-set, open-set, few-shot classification, and domain generalization. Code: https://github.com/katdimitris/Complementary-Dissimilarity-Loss