AOE: Exhaustive Out-of-Distribution Detection via Recalibrating Outlier Labels

2026-05-27

AI Summary

This work addresses a critical limitation in existing out-of-distribution (OOD) detection methods that employ uniform labels for Outlier Exposure, which neglects the latent semantic relationships between OOD samples and in-distribution (ID) classes, leading to suboptimal decision boundaries and overconfident OOD predictions. To overcome this, the authors propose Adaptive-confidence Outlier Exposure (AOE), which reveals for the first time the “over-softening effect” induced by uniform labeling. AOE introduces a learnable temperature parameter to generate adaptive soft targets for outlier samples, thereby preserving semantic structure while increasing prediction entropy. By integrating temperature scaling with soft-target supervision, the method significantly outperforms current state-of-the-art approaches across multiple OOD benchmarks, effectively mitigating overconfidence and enhancing both detection accuracy and robustness.

Abstract

Out-of-distribution (OOD) detection is essential for deploying machine learning models in open-world and safety-critical scenarios, where test inputs may deviate from the training distribution and overconfident predictions on unknown samples can lead to unreliable decisions. Outlier Exposure (OE) has emerged as a promising OOD detection paradigm by introducing auxiliary outliers during training to enlarge the margin between in-distribution (ID) and OOD samples. Existing OE-based methods typically enlarge this margin by employing uniform labels to maximize the entropy of OOD samples over ID categories. However, we theoretically show that uniform labels inevitably disregard the relations between OOD samples and ID categories, termed the over-softening effect, leading to a suboptimal margin bound. Our theoretical analysis further reveals that explicitly exploiting such relations can instead yield improved OOD detection performance. Motivated by this insight, we propose \underline{A}daptive Confidence \underline{OE} (AOE), a simple yet effective method that leverages temperature scaling to recalibrate outlier labels. Specifically, AOE generates adaptive soft targets from temperature-scaled model predictions for OOD samples, where the learnable temperature smooths the prediction distribution without fully erasing class-wise relational information. By supervising OOD samples with these adaptive soft targets, AOE preserves the semantic proximity between OOD samples and ID categories while encouraging the softened targets to approach a high-entropy distribution, thereby suppressing overconfident OOD predictions and enlarging the separation margin. Extensive experiments across diverse benchmarks demonstrate the effectiveness of AOE.

Problem

• Out-of-distribution detection
• Outlier Exposure
• uniform labels
• over-softening effect
• semantic proximity

Innovation

• Out-of-Distribution Detection
• Outlier Exposure
• Temperature Scaling
• Adaptive Soft Labels
• Over-Softening Effect