构建首个多类型网页内容提取基准，覆盖七类网页结构。
WCXB: A Multi-Type Web Content Extraction Benchmark
- 设计五阶段流程生成高质量标注数据，覆盖1613个域名
- 测试13个系统发现文章类准确率高（F1=0.93），其他类型差异大（F1=0.41-0.84）
- 适合研究网页提取、LLM训练与信息抽取的开发者使用
网页内容提取——从页面中分离出主体内容与冗余模板——是搜索索引、检索增强生成、NLP数据集构建和大语言模型训练的前提。现有评估基准受限于规模小（100-800页）、仅覆盖新闻类或基于十余年旧网页。我们提出网页内容提取基准（WCXB），包含2,008个网页，来自1,613个域名，涵盖七种结构不同的页面类型：文章、论坛、产品、集合、列表、文档和服务页。数据集包括1,497页开发集和511页保留测试集，各类别分布均衡。真实标注通过五阶段流程生成：大模型辅助初稿、自动验证、四轮前沿模型评审、片段与质量校验脚本及人工审核。我们评估了13个提取系统（11个启发式、2个神经网络），发现顶级系统在文章类表现优异（F1=0.93），但在结构化页面类型上表现差异显著（F1=0.41-0.84），暴露出现有仅以新闻为主的基准无法发现的盲区。数据集以CC-BY-4.0开源，提供原始HTML、真实标注、页面类型标签和基线结果。
原文摘要 · Abstract (English)
Web content extraction - isolating a page's main content from surrounding boilerplate - is a prerequisite for search indexing, retrieval-augmented generation, NLP dataset construction, and large language model training. Progress in this area has been constrained by the limitations of existing evaluation benchmarks, which are small (100-800 pages), restricted to news articles, or based on web pages from over a decade ago. We introduce the Web Content Extraction Benchmark (WCXB), a dataset of 2,008 web pages from 1,613 domains spanning seven structurally distinct page types: articles, forums, products, collections, listings, documentation, and service pages. The dataset includes a 1,497-page development set and a 511-page held-out test set with matched page type distributions. Ground truth annotations were produced through a five-stage pipeline: LLM-assisted drafting, automated verification, four-pass frontier model review, snippet and quality verification scripts, and human review. We evaluate 13 extraction systems - 11 heuristic and 2 neural - and find that while top systems converge on articles (F1 = 0.93), performance diverges sharply on structured page types (F1 = 0.41-0.84), revealing blind spots invisible to existing article-only benchmarks. The dataset is released under CC-BY-4.0 with HTML source files, ground truth annotations, page type labels, and baseline results.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。