---
title: "WebOverlay: High-Throughput Vision-Only Distillation for Extreme Aspect-Ratio Web Overlay Detection"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2610.41037v1"
permalink: "https://feed.astralor.com/items/itm_495bec28e0543558f1055972"
published: "2026-10-29T22:29:33.000Z"
score: 55
channel: "tech"
tags: ["LLM","推理效率","多模态","WebOverlay","High-Throughput Vision-Only Distillation","Extreme Aspect-Ratio Web","Overlay Detection Web","While"]
---

# WebOverlay: High-Throughput Vision-Only Distillation for Extreme Aspect-Ratio Web Overlay Detection

## At a glance

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…

## Key points

- 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.

## Full text

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.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_495bec28e0543558f1055972