arXiv cs.AI· Boyu Niu, Yingjie Xiong, Shurui Li, Linhan Pang, Yuntong Zhang, Han Wang, Hao Wu, Kai Zuo, Xu Tang, Xuanhe Zhou, Fan Wu·· 11 小时前AI 评分34
AmbiHTMLBench：面向 HTML 视觉丰富文档解析的基准测试
AmbiHTMLBench: A Benchmark for HTML-based Visually Rich Document parsing
AI 导读
研究者提出 AmbiHTMLBench，用于评测模型从真实文档还原可执行 HTML 的能力，要求保留内容、组织与视觉外观。该基准包含 1,115 个多样化页面，采用客观协议（表格保真度、页面级文本误差、二维布局一致性）与参考锚定的 LLM 评判协议双重评估。在十个多模态系统上，最佳客观得分为 0.8092，但主观胜出者不同；微调的小参数专家模型在文档转 HTML 重建任务上表现良好。
正文 · 原文
Authors:Boyu Niu, Yingjie Xiong, Shurui Li, Linhan Pang, Yuntong Zhang, Han Wang, Hao Wu, Kai Zuo, Xu Tang, Xuanhe Zhou, Fan Wu
Abstract:Traditional document parsing systems use Markdown as their output representation, which is effective for linear text but does not fully preserve fine-grained spatial structure, table geometry, visual hierarchy, or rendering semantics. We show that HTML is advantageous for heterogeneous and layout-sensitive documents whose representations must remain editable, inspectable, and reusable, and introduce AmbiHTMLBench to systematically study this setting. In this benchmark, given a real-world document, a model must recover executable HTML that preserves its content, organization, and visual appearance. The benchmark contains 1,115 diverse pages with structured annotations and evaluates predictions in two different ways: The objective protocol aligns ground-truth and predicted rendered blocks and measures table fidelity, page-level text error, and a two-dimensional layout consistency; a complementary reference-anchored protocol measures perceived fidelity by aligning LLM judges with human preferences. Across ten primary multimodal systems, the best objective score is 0.8092, while the subjective winner is different, demonstrating that automatic decomposition and human-perceived fidelity capture complementary properties. We further fine-tune a small-parameter expert model and find that task-specific adaptation yields good performance on document-to-HTML reconstruction task. These results position AmbiHTMLBench as a diagnostic testbed for building smarter and more logical document parsing systems.
| Comments: | Check latest benchmark info on this https URL |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.06449 [cs.AI] |
| (or arXiv:2611.06449v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.06449 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Boyu Niu [view email]
[v1]
Wed, 4 Nov 2026 12:51:26 UTC (28,912 KB)
来源：arXiv cs.AI · arxiv.org