{
  "abstract": "Industrial surface anomaly detection under data-scarce, unsupervised conditions requires models that are both label-efficient and computationally lightweight. This paper presents LKDF-AD, a knowledge-distillation framework combining a frozen pretrained teacher with a compact student network (385,168 parameters), trained using only k = 1–10 normal reference images per category, without any labeled anomalies. Evaluated across 15 MVTec AD categories and five shot counts, the framework achieves 77.88–86.19% image-level and 90.59–94.87% pixel-level AUROC, and generalizes to the independent BTAD benchmark (73.6–90.99% image-level AUROC), with a larger BTAD pixel-AUROC/PRO gap flagged as an open question. Measured end-to-end (teacher plus student together, as deployed), the framework uses 1.93× fewer parameters (11.56 M vs. 22.35 M) and achieves 1.57× lower inference latency (4.436 ms vs. 6.954 ms) than a full-size STFPM-style baseline trained under an identical protocol. Under category-level paired testing (using 15 independent category means, the statistically appropriate unit given repeated measurements across shot counts), neither the image-level nor the pixel-level advantage over STFPM reaches statistical significance (two-sided p = 0.7197 and p = 0.188, respectively). Consequently, the framework’s established advantage over STFPM is its directly measured computational efficiency (1.93× fewer end-to-end parameters, 1.57× lower end-to-end latency), rather than accuracy. In contrast, against the RegAD-inspired baseline, both accuracy advantages remain highly significant under the same corrected test (two-sided p = 0.00201 for image-level and p = 0.00153 for pixel-level). A random, completely untrained student achieves only near-chance performance (52.7% image-level, 46.6% pixel-level AUROC), confirming that the observed localization signal reflects genuine distillation rather than an architectural artifact. A dedicated investigation of the framework’s weakest category (screw, near-chance performance) tested a rotation-sensitivity hypothesis directly across the full test set; the result was a significant correlation in the direction opposite to that hypothesized, and we report the mechanism as unresolved rather than settled. These results position the framework as a genuinely lightweight, empirically characterized — rather than definitively superior — approach to few-shot unsupervised industrial anomaly detection.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": null,
  "author_total": 1,
  "authors": [
    {
      "name": "Xian Gao"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/64mmbjzp89shh.bib",
  "cited_by": 0,
  "doi": "10.1038/s41598-026-74501-8",
  "fulltext_url": null,
  "id": "64mmbjzp89shh",
  "issns": [
    "2045-2322"
  ],
  "landing_url": "https://www.nature.com/articles/s41598-026-74501-8",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0",
  "metadata_licence": null,
  "native_subjects": [
    [
      "openalex",
      "2209|Industrial Vision Systems and Defect Detection"
    ]
  ],
  "openalex": "W7222296774",
  "pdf_url": "https://www.nature.com/articles/s41598-026-74501-8_reference.pdf",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-15",
  "publisher": "Springer Science and Business Media LLC",
  "ris": "https://research.draconitic.com/paper/64mmbjzp89shh.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:17:28Z",
      "native_id": "10.1038/s41598-026-74501-8",
      "source": "crossref",
      "url": "https://www.nature.com/articles/s41598-026-74501-8"
    }
  ],
  "subjects": [],
  "title": "A lightweight knowledge distillation framework for few shot unsupervised industrial surface anomaly detection and localization",
  "title_text": "A lightweight knowledge distillation framework for few shot unsupervised industrial surface anomaly detection and localization",
  "type": "peer-reviewed",
  "updated_at": "2026-10-30T12:17:28Z",
  "url": "https://research.draconitic.com/paper/64mmbjzp89shh",
  "venue": "Scientific Reports",
  "versions": []
}