Spurious-Aware Prototype Refinement for Reliable Out-of-Distribution Detection
This addresses reliability and safety issues for machine learning models in real-world applications by improving robustness against spurious correlations in OOD detection.
The paper tackles the problem of out-of-distribution (OOD) detection being vulnerable to spurious correlations, proposing SPROD, a prototype-based method that refines class prototypes to mitigate bias, resulting in average improvements of 4.8% in AUROC and 9.4% in FPR@95 over the second best method.
Out-of-distribution (OOD) detection is crucial for ensuring the reliability and safety of machine learning models in real-world applications, where they frequently face data distributions unseen during training. Despite progress, existing methods are often vulnerable to spurious correlations that mislead models and compromise robustness. To address this, we propose SPROD, a novel prototype-based OOD detection approach that explicitly addresses the challenge posed by unknown spurious correlations. Our post-hoc method refines class prototypes to mitigate bias from spurious features without additional data or hyperparameter tuning, and is broadly applicable across diverse backbones and OOD detection settings. We conduct a comprehensive spurious correlation OOD detection benchmarking, comparing our method against existing approaches and demonstrating its superior performance across challenging OOD datasets, such as CelebA, Waterbirds, UrbanCars, Spurious Imagenet, and the newly introduced Animals MetaCoCo. On average, SPROD improves AUROC by 4.8% and FPR@95 by 9.4% over the second best.