arXiv cs.AI· Donghwan Kim, Youngwoong Jin, Hyunsoo Yoon·· 12 小时前AI 评分31
反一致性正则化绝对缩放：面向分布外检测的 SCALE 改进方法
Anti-coherence Regularized Absolute Scaling for Out-of-Distribution Detection
AI 导读
针对 SCALE 在负激活存在时幅度比定义失效、在 ViT 等骨干网络上退化的问题，研究者提出 Absolute Fix，改用绝对值激活计算该比例，并在此基础上引入 Anti-coherence 正则化，按样本对最大 logit 贡献中符号不匹配的比例惩罚缩放因子。
正文 · 原文
Abstract:Out-of-distribution (OOD) detection is recognized as a critical problem in AI safety. SCALE is an effective post-hoc OOD detection method in the activation-shaping family that requires no training data and rescales penultimate activations using a percentile-based magnitude ratio. However, this ratio is ill-defined when negative activations are present, since positive and negative contributions cancel within the underlying sum, causing SCALE to degenerate on backbones such as ViT. We address this with an Absolute Fix that computes the ratio over absolute-valued activations rather than discarding negative activations through ReLU. Building on this corrected scale, we propose Anti-coherence regularization, which penalizes the scaling factor by the fraction of a sample's max-logit contributions to the max logit that are sign-mismatched and thereby suppresses unreliable evidence for the predicted class. Despite requiring no access to training data or training labels, our method ties or outperforms activation-shaping methods that rely on training data in terms of AUROC and FPR@95 on CIFAR-10 and ImageNet under the OpenOOD v1.5 protocol.
| Comments: | 29 pages, 6 figures |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.32255 [cs.LG] |
| (or arXiv:2610.32255v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.32255 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Donghwan Kim [view email]
[v1]
Fri, 23 Oct 2026 07:09:51 UTC (4,220 KB)
来源：arXiv cs.AI · arxiv.org