通过解耦特征与置信度，提升模型对分布外样本的检测能力。
GAFD-CC: Global-Aware Feature Decoupling with Confidence Calibration for OOD Detection
- 基于分类权重全局解耦特征，分离正负相关成分
- 融合多尺度置信度，显著降低误报率
- 无需重训练，适合实际部署场景
分布外（OOD）检测对于保障学习模型在真实应用中的可靠性与鲁棒性至关重要。现有后处理型OOD检测方法依赖特征与预测概率信息进行检测，无需重新训练，但常忽略特征与输出之间内在关联，影响检测效果。为此，本文提出全局感知特征解耦与置信度校准方法（GAFD-CC）。该方法首先利用分类权重引导的全局感知机制，将特征对齐至全局分类方向，实现特征解耦，从中提取两类关键信息：促进分布内/外边界优化的正相关特征，以及抑制误报、收紧边界的负相关特征。其次，自适应融合解耦特征与多尺度基于逻辑值的置信度，实现全面且稳健的OOD检测。在大规模基准上的大量实验表明，GAFD-CC性能优于当前最优方法，具备强泛化能力。
原文摘要 · Abstract (English)
Out-of-distribution (OOD) detection is paramount to ensuring the reliability and robustness of learning models in real-world applications. Existing post-hoc OOD detection methods detect OOD samples by leveraging their features and logits information without retraining. However, they often overlook the inherent correlation between features and logits, which is crucial for effective OOD detection. To address this limitation, we propose Global-Aware Feature Decoupling with Confidence Calibration (GAFD-CC). GAFD-CC aims to refine decision boundaries and increase discriminative performance. Firstly, it performs global-aware feature decoupling guided by classification weights. This involves aligning features with the direction of global classification weights to decouple them. From this, GAFD-CC extracts two types of critical information: positively correlated features that promote in-distribution (ID)/OOD boundary refinement and negatively correlated features that suppress false positives and tighten these boundaries. Secondly, it adaptively fuses these decoupled features with multi-scale logit-based confidence for comprehensive and robust OOD detection. Extensive experiments on large-scale benchmarks demonstrate GAFD-CC's competitive performance and strong generalization ability compared to those of state-of-the-art methods.
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。