Safety
A Systematic Comparison of Training Objectives for Out-of-Distribution Detection in Image Classification
arXiv:2603.07571v3 Announce Type: replace-cross Abstract: Out-of-distribution (OOD) detection is critical in safety-sensitive applications. While this challenge has been addressed from various perspec
arXiv:2603.07571v3 Announce Type: replace-cross Abstract: Out-of-distribution (OOD) detection is critical in safety-sensitive applications. While this challenge has been addressed from various perspectives, the influence of training objectives on OOD behavior remains comparatively underexplored. In this paper, we present a systematic comparison of four widely used training objectives: Cross-Entropy Loss, Prototype Loss, Triplet Loss, and Average Precision (AP) Loss, spanning probabilistic, prototype-based, metric-learning, and ranking-based supervision, for OOD detection in image classification under standardized OpenOOD protocols. Within the evaluated ResNet-18/OpenOOD setting and objective-specific OOD scoring rules, Cross-Entropy Loss, Prototype Loss, and AP Loss achieve comparable in-distribution accuracy, while Cross-Entropy Loss provides the most consistent near- and far-OOD AUROC overall; the other objectives can be competitive in specific settings.
Related
- Masked Diffusion Modeling for Anomaly Detection
- From Out-of-Distribution Detection to Hallucination Detection: A Geometric View
- Beyond Penalization: Diffusion-based Out-of-Distribution Detection and Selective Regularization in Offline Reinforcement Learning
Source: arXiv cs.AI | 2026-08-17