无需标注数据，自动识别关键特征层实现跨尺度的异常检测
MM++: Unsupervised Scale-Invariant Multilayer OOD Detection via Top-K Gated Feature Fusion
- 通过熵密度下降筛选关键中间层，融合多层特征构建联合表示
- 在不依赖外部数据下，对近域和远域异常均保持高检测准确率
- 适合无监督场景下的模型可靠性评估，尤其适用于部署前验证
我们提出MM++（Multilayer Mahalanobis++），一种完全无监督、严格后处理且尺度不变的分布外（OOD）检测框架。为解决尺度不变性与层级表达力之间的权衡，MM++构建了一个理论严谨的联合特征空间。首先通过测量熵密度下降来识别具有判别性的中间层，这些层标志着语义压缩的突变边界。将选定层与终端表示融合，捕获潜在的跨层相关性，同时抑制早期层噪声。关键的是，采用Ledoit-Wolf正则化的共享协方差矩阵稳定了该统一空间，实现可靠的距离估计。无需辅助的OOD数据、分类器微调或架构修改，MM++在不同架构上对近域和远域OOD检测均表现出稳健性能。
原文摘要 · Abstract (English)
We introduce MM++ (Multilayer Mahalanobis++), a fully unsupervised, strictly post-hoc, and scale-invariant framework for Out-of-Distribution (OOD) detection. To address the trade-off between scale invariance and hierarchical expressivity, MM++ constructs a principled joint feature space. It first identifies discriminative intermediate layers by measuring entropy density drops, which mark the boundaries of sharp semantic compression. By fusing these selected layers with the terminal representation, the framework captures latent cross-layer correlations while mitigating early-layer noise. Crucially, a Ledoit-Wolf regularized tied covariance matrix stabilizes this unified space, enabling reliable distance estimation. Requiring no auxiliary OOD data, classifier fine-tuning, or architectural modifications, MM++ delivers robust performance across distinct architectures for both near- and far-OOD detection.
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。