Metraipolis › archive › dc_c012e1e389be

PL-SCEA: Reconfiguring Pretrained Attention for Few-Shot Industrial Anomaly Detection — arXiv:2609.03655

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.03655
License (as archived)
CC0-1.0
Archived
2026-10-17T02:56:23.633Z
SHA-256 of the archived text
dbb6349ed42ed072ca65a4a2e2d6c2449de66631cba3a61fbbe95530d68adf4a
Stable URL of this page
https://metraipolis.com/documento/dc_c012e1e389be

How to cite

PL-SCEA: Reconfiguring Pretrained Attention for Few-Shot Industrial Anomaly Detection — arXiv:2609.03655 — arXiv (en), https://arxiv.org/abs/2609.03655 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c012e1e389be (Metraipolis archive dc_c012e1e389be, sha256 dbb6349ed42ed072ca65a4a2e2d6c2449de66631cba3a61fbbe95530d68adf4a)
BibTeX
@misc{dcc012e1e389be,
  title = {PL-SCEA: Reconfiguring Pretrained Attention for Few-Shot Industrial Anomaly Detection — arXiv:2609.03655},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.03655},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c012e1e389be; sha256 dbb6349ed42ed072ca65a4a2e2d6c2449de66631cba3a61fbbe95530d68adf4a},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.03655. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision Foundation Models (VFMs) provide transferable patch representations for few-shot industrial anomaly detection, but their attention computation is typically inherited from pretraining objectives centered on semantic aggregation. This creates a potential mismatch: token relations that support semantic recognition may not adequately expose the localized texture and structural deviations required for anomaly localization. We therefore investigate the hypothesis that the attention computation of a frozen VFM can be reconfigured as a task-relevant component of anomaly detection. We instantiate this idea with Power-Law Self-Correlation Enhanced Attention (PL-SCEA), which retains the semantic context of pretrained query-key attention while constructing token-adaptive self-correlations over contextualized value features. Positive-correlation filtering and power-law reweighting then emphasize relations that are salient relative to each token's relational background, without introducing additional trainable attention projections. The resulting features are modeled by a lightweight variational autoencoder that provides a fixed-size reconstruction-based representation of category-specific normality. The two stages serve complementary roles: attention reconfiguration shapes how local relational deviations are represented, while reconstruction-based modeling converts deviations from learned normality into anomaly scores. Across MVTec AD and VisA, the complete framework achieves competitive image-level detection and consistently strong pixel-level localization across the evaluated few-shot settings. Ablations further show that PL-SCEA improves localization with either the VAE or a memory bank under the tested setting. These results support the view that task-aligned attention reconfiguration can improve the anomaly-localization capability of frozen pretrained representations.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-04 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.03655 — 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 2026-10-17T02:27:18.979Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-04?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).