Metraipolis › archive › dc_e6d77d23ca4b
Fault-Guided Cross-Layer Prediction for Unsupervised Anomaly Detection — arXiv:2611.08140
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.08140
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-06T14:53:43.441Z
- SHA-256 of the archived text
685b213334052143589e03b2dbdbd3aa61d79f5d9802ac5d82e8faeff03f50d5- Stable URL of this page
https://metraipolis.com/documento/dc_e6d77d23ca4b
How to cite
Fault-Guided Cross-Layer Prediction for Unsupervised Anomaly Detection — arXiv:2611.08140 — arXiv (en), https://arxiv.org/abs/2611.08140 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e6d77d23ca4b (Metraipolis archive dc_e6d77d23ca4b, sha256 685b213334052143589e03b2dbdbd3aa61d79f5d9802ac5d82e8faeff03f50d5)
BibTeX
@misc{dce6d77d23ca4b,
title = {Fault-Guided Cross-Layer Prediction for Unsupervised Anomaly Detection — arXiv:2611.08140},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.08140},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e6d77d23ca4b; sha256 685b213334052143589e03b2dbdbd3aa61d79f5d9802ac5d82e8faeff03f50d5},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.08140. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Feature reconstruction is widely used for unsupervised industrial anomaly detection in the single-class setting, where only normal images from one product category are available for training. We propose a single-class detector that asks what information one block can pass to another. We treat the depth of a frozen vision transformer as a sequence of discrete transitions. A directed predictor learns a block-to-block hidden-state transition pointwise over tokens. Adjacent-layer cosine similarity exposes four sharp fault lines in DINOv2 ViT-L, separating stable depth segments. The deepest blocks retain semantic context but lose local detail; a prediction from deep or shallow sources to sensitive targets therefore exposes the information that cannot be transported. The fault structure fixes the source and target roles, and the same rule transfers to ViT-B and RADIO-B. We use a bounded foreground prior calibrated solely on training images: it attenuates background evidence without also erasing defects outside the estimated foreground. Using one predictor architecture and a common training-only calibration protocol, image AUROC is 99.75 on MVTec-AD, 99.42 on VisA and 94.27 on Real-IAD. On Real-IAD the object-level AUROC is 96.26. At 256x256 pixel resolution, P-AUROC/P-AUPRO are 98.46/95.97 on MVTec-AD, 99.21/96.12 on VisA and 99.26/96.59 on Real-IAD. Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2611.08140 — 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 2026-11-06T14:25:42.626Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).