用排序损失提升分布外检测，性能超越当前最佳。
RankOOD -- Class Ranking-based Out-of-Distribution Detection
- 基于类别排序构建新检测框架，利用排列概率判断异常。
- 在TinyImageNet近似分布外任务上FPR95降低4.3%。
- 适合需要高精度异常识别的工业部署场景。
我们提出RankOOD，一种基于排序的分布外（OOD）检测方法，通过在基础模型中广泛使用的Plackett-Luce损失进行训练。该方法的核心洞察是：使用交叉熵损失训练的深度模型，在分布内（ID）类别预测时会自然形成类别排名模式。RankOOD框架首先利用初始分类器为每个类别提取排名列表，再通过第二轮训练以Plackett-Luce损失优化，将类别固定排列作为预测目标。对于分布外样本，虽然可能被高概率分配给某个分布内样本，但其满足该类别排序规律的概率通常很低。在近似分布外的TinyImageNet评估基准上，RankOOD实现了当前最优性能，使FPR95降低了4.3%。
原文摘要 · Abstract (English)
We propose RankOOD, a rank-based Out-of-Distribution (OOD) detection approach based on training a model with the Placket-Luce loss, which is now extensively used for preference alignment tasks in foundational models. Our approach is based on the insight that with a deep learning model trained using the Cross Entropy Loss, in-distribution (ID) class prediction induces a ranking pattern for each ID class prediction. The RankOOD framework formalizes the insight by first extracting a rank list for each class using an initial classifier and then uses another round of training with the Plackett-Luce loss, where the class rank, a fixed permutation for each class, is the predicted variable. An OOD example may get assigned with high probability to an ID example, but the probability of it respecting the ranking classification is likely to be small. RankOOD, achieves SOTA performance on the near-ODD TinyImageNet evaluation benchmark, reducing FPR95 by 4.3%.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。