Metraipolis › archive › dc_7ee2d0bd3a48

WebOverlay: High-Throughput Vision-Only Distillation for Extreme Aspect-Ratio Web Overlay Detection — arXiv:2610.41037

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.41037
License (as archived)
CC0-1.0
Archived
2026-11-02T07:10:26.373Z
SHA-256 of the archived text
7178990dbe6ea8c33ca969880638be2360a08bcc50117725ad301283dbb348d2
Stable URL of this page
https://metraipolis.com/documento/dc_7ee2d0bd3a48

How to cite

WebOverlay: High-Throughput Vision-Only Distillation for Extreme Aspect-Ratio Web Overlay Detection — arXiv:2610.41037 — arXiv (en), https://arxiv.org/abs/2610.41037 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7ee2d0bd3a48 (Metraipolis archive dc_7ee2d0bd3a48, sha256 7178990dbe6ea8c33ca969880638be2360a08bcc50117725ad301283dbb348d2)
BibTeX
@misc{dc7ee2d0bd3a48,
  title = {WebOverlay: High-Throughput Vision-Only Distillation for Extreme Aspect-Ratio Web Overlay Detection — arXiv:2610.41037},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.41037},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7ee2d0bd3a48; sha256 7178990dbe6ea8c33ca969880638be2360a08bcc50117725ad301283dbb348d2},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41037. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Web overlays, such as cookie consent banners, newsletter prompts, and age gates, obstruct automated web crawlers and degrade downstream document representations. While proprietary multimodal Large Language Models (MLLMs) can accurately identify overlays, their latency and inference cost preclude web-scale ingestion. Conversely, lightweight vision models struggle with full-page renderings characterized by extreme vertical aspect ratios, where standard isotropic resizing collapses critical text features and causes false positives. In this work, we present WebOverlay, a high-throughput vision-only detector trained via distillation from a frontier vision-language teacher. We establish a dynamic tiling strategy that partitions extreme aspect-ratio web renders into aspect-preserving patches with spatial overlap, coupled with an aspect-ratio-aware boundary adjudication mechanism to eliminate boundary artifacts. To eliminate cross-modal latency overheads, our student model relies purely on visual cues, bypassing DOM tree parsing. WebOverlay significantly outperforms DOM-based heuristics and achieves competitive precision and recall against heavy proprietary MLLM baselines while delivering multiple orders of magnitude higher throughput and substantially reduced deployment costs.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI); Information Retrieval (cs.IR).
Pagina del lavoro: https://arxiv.org/abs/2610.41037 — 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 2026-11-02T07:06:59.415Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).