论文标题
vim：与虚拟逻辑匹配的分发
ViM: Out-Of-Distribution with Virtual-logit Matching
论文作者
论文摘要
大多数现有的分布（OOD）检测算法都取决于单个输入来源：功能，logit或SoftMax概率。但是，OOD示例的巨大多样性使这种方法脆弱。有一些OOD样品在特征空间中易于识别，而很难区分在逻辑空间中，反之亦然。在这个观察结果的激励下，我们提出了一种名为Virtual-Logit匹配（VIM）的新颖的OOD评分方法，该方法结合了特征空间中的类别不稳定分数和分布分布（ID）类依赖类ligits的分数。具体而言，代表虚拟OOD类的附加logit是由功能的残差与主空间生成的，然后通过恒定缩放与原始logits匹配。 SoftMax后此虚拟logit的概率是OOD度的指标。为了促进对学术界的大规模OOD检测的评估，我们为Imagenet-1k创建了一个新的OOD数据集，该数据集已被人类通知，是现有数据集的大小的8.8倍。我们进行了广泛的实验，包括CNN和视觉变压器，以证明拟议的VIM评分的有效性。特别是，使用BIT-S模型，我们的方法在四个困难的OOD基准测试中获得平均AUROC 90.91％，比最佳基线领先4％。代码和数据集可从https://github.com/haoqiwang/vim获得。
Most of the existing Out-Of-Distribution (OOD) detection algorithms depend on single input source: the feature, the logit, or the softmax probability. However, the immense diversity of the OOD examples makes such methods fragile. There are OOD samples that are easy to identify in the feature space while hard to distinguish in the logit space and vice versa. Motivated by this observation, we propose a novel OOD scoring method named Virtual-logit Matching (ViM), which combines the class-agnostic score from feature space and the In-Distribution (ID) class-dependent logits. Specifically, an additional logit representing the virtual OOD class is generated from the residual of the feature against the principal space, and then matched with the original logits by a constant scaling. The probability of this virtual logit after softmax is the indicator of OOD-ness. To facilitate the evaluation of large-scale OOD detection in academia, we create a new OOD dataset for ImageNet-1K, which is human-annotated and is 8.8x the size of existing datasets. We conducted extensive experiments, including CNNs and vision transformers, to demonstrate the effectiveness of the proposed ViM score. In particular, using the BiT-S model, our method gets an average AUROC 90.91% on four difficult OOD benchmarks, which is 4% ahead of the best baseline. Code and dataset are available at https://github.com/haoqiwang/vim.
