Sign inGet started
Paper · arXiv

Backbone-Equated Diffusion OOD via Sparse Internal Snapshots

Fair comparison between diffusion-based OOD detectors is challenging, as conclusions can vary with backbone choice, corruption parameterization, and test-time budget. We address this issue through a Mutualized Backbone-Equated (MBE) protocol that aligns canonical corruption levels and logical test-time cost across diffusion backbones. Within this setting, we introduce Canonical Feature Snapshots (CFS), a family of detectors that probes a frozen diffusion backbone using only a tiny number of native internal activations at canonical low-noise levels. On a controlled CIFAR-scale benchmark, the strongest one-forward CFS variant is CFS(1x2), while an even smaller decoder-only variant remains h...

AuthorsYadang Alexis Rouzoumka, Jean Pinsolle, Eugénie Terreaux et al.
Primary categorycs.LG
Published10 May 2026
Last updated at source10 May 2026
cs.LGcs.AI
Embed this entry
<iframe src="https://silytics.si/embed/item/d81eca82-3efd-43c5-b22a-6330e61b5336" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.