Metraipolis › archive › dc_ff53541dcc54
Layer Selection in VLMs for Zero-Shot OOD Detection via Multi-Resolution Entropy Estimation — arXiv:2609.08524
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.08524
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T15:12:19.173Z
- SHA-256 of the archived text
95342c2bd7bb2a8cfb79a8089c8abe8b4228ab641aa28f3e7117a37e118b3b87- Stable URL of this page
https://metraipolis.com/documento/dc_ff53541dcc54
How to cite
Layer Selection in VLMs for Zero-Shot OOD Detection via Multi-Resolution Entropy Estimation — arXiv:2609.08524 — arXiv (en), https://arxiv.org/abs/2609.08524 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ff53541dcc54 (Metraipolis archive dc_ff53541dcc54, sha256 95342c2bd7bb2a8cfb79a8089c8abe8b4228ab641aa28f3e7117a37e118b3b87)
BibTeX
@misc{dcff53541dcc54,
title = {Layer Selection in VLMs for Zero-Shot OOD Detection via Multi-Resolution Entropy Estimation — arXiv:2609.08524},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.08524},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ff53541dcc54; sha256 95342c2bd7bb2a8cfb79a8089c8abe8b4228ab641aa28f3e7117a37e118b3b87},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.08524. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Out-of-distribution (OOD) detection is crucial for safe deployment of medical AI systems, where domain shifts arise across institutions, acquisition protocols, and patient populations. VLMs enable zero-shot OOD detection by embedding images into a language-aligned latent space, where cross-modal similarity serves as a non-parametric confidence signal for identifying in-distribution samples. Yet existing methods rely almost exclusively on final-layer embeddings, implicitly assuming that the deepest representations are universally optimal. We first show that this assumption does not hold in medical imaging: intermediate layers provide complementary OOD signals, and the optimal representational depth depends on the respective image modality. While prior work selects layer combinations via entropy minimization of normalized histograms, we demonstrate that single-resolution entropy estimation is highly sensitive to binning choices, leading to performance variations of up to 19.3% AUROC. To address this instability, we propose a multi-resolution entropy estimation strategy that aggregates histogram statistics across multiple discretization scales, enabling robust and stable intermediate-layer selection. Across two medical OOD benchmarks, namely MIDOG and OASIS, covering distinct imaging modalities, diverse shift types, and different VLM backbones, our method consistently outperforms state-of-the-art approaches, offering a lightweight and stable solution for zero-shot OOD detection. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.08524 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-16T14:49:57.324Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).