发现分类分裂异常检测中评分方向不稳定问题，影响模型可靠性。
Testing the Test: Score-Direction Instability in Class-Split Anomaly Detection
- 提出无训练诊断方法：邻域类别泄露，检测评分不稳定性。
- 在多个数据集上验证，评分可能趋近随机甚至反转。
- 适合关注异常检测评估可信度的研究者阅读。
类分裂异常检测的内部数据集评估广泛用作无条件分布外检测的代理。我们发现，当被排除的异常类别在表示空间中与正常混合重叠时，该评估协议可能失效。在此情况下，异常评分可能退化至随机水平甚至反转，且最优评分方向依赖于未知的异常类别。本文提出一种简单无训练的诊断方法——邻域类别泄露，并在Fashion-MNIST、CIFAR-10和Imagenette数据集上验证其对像素空间和VAE潜在空间的评分方向不稳定性预测能力。结果表明，类分裂异常检测基准应被视为几何依赖性压力测试，而非异常检测能力的无条件证据。
原文摘要 · Abstract (English)
Within-dataset class-split evaluation is widely used as a proxy for fully unconditional out-of-distribution anomaly detection. We show that this protocol can become ill-posed when the held-out anomaly class overlaps the normal mixture in representation space. In this regime, anomaly scores may collapse toward chance or even invert, and the preferred score direction can depend on the unknown anomaly class. We introduce a simple training-free diagnostic, neighborhood class leakage, and show that it predicts score-direction instability across Fashion-MNIST, CIFAR-10, and Imagenette, in both pixel and VAE latent spaces. Our results suggest that class-split AD benchmarks should be treated as geometry-dependent stress tests rather than unconditional evidence of anomaly-detection ability.
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。