Early High-Frequency Injection for Geometry-Sensitive OOD Detection

TL;DR

This paper introduces EIHF, a high-frequency input intervention that enhances geometry-sensitive out-of-distribution detection by exposing high-frequency evidence early, improving detection performance on CIFAR-100 and ImageNet-100.

Contribution

The paper proposes EIHF, a novel input-side intervention that leverages high-frequency information to improve geometry-sensitive OOD detection without retraining models.

Findings

EIHF improves OOD detection performance on CIFAR-100.

EIHF achieves the best average FPR95 on ImageNet-100.

EIHF reveals limitations on scene-centric dataset shifts.

Abstract

Post-hoc OOD detectors score logits or features after training, so their success depends on the geometry already encoded in the representation. We revisit this assumption through a band-wise MMD^2 analysis across CE, SimCLR, SupCon, and the OOD-oriented representation method PALM. In our diagnostic, low-frequency input bands induce weaker ID/OOD feature discrepancy, whereas higher-frequency bands tend to provide stronger separability. This observation motivates EIHF, an input-side intervention that exposes high-frequency evidence before the first convolution without changing the training objective. EIHF is strongest for geometry-sensitive OOD detection: under matched training and scoring settings, it reshapes class-conditional feature geometry and reduces ID/OOD Mahalanobis score overlap. Experiments on CIFAR-100 and ImageNet-100 show gains on CIFAR-100 and the best average FPR95 with second-best average AUROC on ImageNet-100, while also revealing a limitation on the scene-centric Places shift. Code is available at https://anonymous.4open.science/r/EIHF.