MosaicDoc: A Large-Scale Bilingual Benchmark for Visually Rich Document Understanding

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Chen, Ketong, Chen, Yuhao, Xue, Yang
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908650213212160
author Chen, Ketong
Chen, Yuhao
Xue, Yang
author_facet Chen, Ketong
Chen, Yuhao
Xue, Yang
contents Despite the rapid progress of Vision-Language Models (VLMs), their capabilities are inadequately assessed by existing benchmarks, which are predominantly English-centric, feature simplistic layouts, and support limited tasks. Consequently, they fail to evaluate model performance for Visually Rich Document Understanding (VRDU), a critical challenge involving complex layouts and dense text. To address this, we introduce DocWeaver, a novel multi-agent pipeline that leverages Large Language Models to automatically generate a new benchmark. The result is MosaicDoc, a large-scale, bilingual (Chinese and English) resource designed to push the boundaries of VRDU. Sourced from newspapers and magazines, MosaicDoc features diverse and complex layouts (including multi-column and non-Manhattan), rich stylistic variety from 196 publishers, and comprehensive multi-task annotations (OCR, VQA, reading order, and localization). With 72K images and over 600K QA pairs, MosaicDoc serves as a definitive benchmark for the field. Our extensive evaluation of state-of-the-art models on this benchmark reveals their current limitations in handling real-world document complexity and charts a clear path for future research.
format Preprint
id arxiv_https___arxiv_org_abs_2511_09919
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MosaicDoc: A Large-Scale Bilingual Benchmark for Visually Rich Document Understanding
Chen, Ketong
Chen, Yuhao
Xue, Yang
Computer Vision and Pattern Recognition
Despite the rapid progress of Vision-Language Models (VLMs), their capabilities are inadequately assessed by existing benchmarks, which are predominantly English-centric, feature simplistic layouts, and support limited tasks. Consequently, they fail to evaluate model performance for Visually Rich Document Understanding (VRDU), a critical challenge involving complex layouts and dense text. To address this, we introduce DocWeaver, a novel multi-agent pipeline that leverages Large Language Models to automatically generate a new benchmark. The result is MosaicDoc, a large-scale, bilingual (Chinese and English) resource designed to push the boundaries of VRDU. Sourced from newspapers and magazines, MosaicDoc features diverse and complex layouts (including multi-column and non-Manhattan), rich stylistic variety from 196 publishers, and comprehensive multi-task annotations (OCR, VQA, reading order, and localization). With 72K images and over 600K QA pairs, MosaicDoc serves as a definitive benchmark for the field. Our extensive evaluation of state-of-the-art models on this benchmark reveals their current limitations in handling real-world document complexity and charts a clear path for future research.
title MosaicDoc: A Large-Scale Bilingual Benchmark for Visually Rich Document Understanding
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.09919