Ranked Activation Shift for Post-Hoc Out-of-Distribution Detection
For practitioners needing reliable OOD detection, this method offers a simple, plug-and-play solution that works consistently across models and datasets without hyperparameter tuning.
The paper identifies instability in existing post-hoc OOD detection methods due to activation distribution differences and proposes a hyperparameter-free method that replaces sorted activation magnitudes with a fixed in-distribution reference profile, achieving strong and consistent performance across datasets and architectures without tuning.
State-of-the-art post-hoc out-of-distribution detection methods rely on intermediate layer activation editing. However, they exhibit inconsistent performance across datasets and models. We show that this instability is driven by differences in the activation distributions, and identify a failure mode of scaling-based methods that arises when penultimate layer activations are not rectified. Motivated by this analysis, we propose \ours, a hyperparameter-free post-hoc method that replaces sorted activation magnitudes with a fixed in-distribution reference profile. Our simple plug-and-play method shows strong and consistent performance across datasets and architectures without assumptions on the penultimate layer activation function, and without requiring any hyperparameter tuning, while preserving in-distribution classification accuracy by construction. We further analyze what drives the improvement, showing that both inhibiting and exciting activation shifts independently contribute to better out-of-distribution discrimination.