Saved in:
Bibliographic Details
Main Authors: Liu, Shuai, Li, Youmeng, Wei, Jizeng
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.10258
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910002680168448
author Liu, Shuai
Li, Youmeng
Wei, Jizeng
author_facet Liu, Shuai
Li, Youmeng
Wei, Jizeng
contents Document Reading Order Recovery is a fundamental task in document image understanding, playing a pivotal role in enhancing Retrieval-Augmented Generation (RAG) and serving as a critical preprocessing step for large language models (LLMs). Existing methods often struggle with complex layouts(e.g., multi-column newspapers), high-overhead interactions between cross-modal elements (visual regions and textual semantics), and a lack of robust evaluation benchmarks. We introduce XY-Cut++, an advanced layout ordering method that integrates pre-mask processing, multi-granularity segmentation, and cross-modal matching to address these challenges. Our method significantly enhances layout ordering accuracy compared to traditional XY-Cut techniques. Specifically, XY-Cut++ achieves state-of-the-art performance (98.8 BLEU overall) while maintaining simplicity and efficiency. It outperforms existing baselines by up to 24\% and demonstrates consistent accuracy across simple and complex layouts on the newly introduced DocBench-100 dataset. This advancement establishes a reliable foundation for document structure recovery, setting a new standard for layout ordering tasks and facilitating more effective RAG and LLM preprocessing.
format Preprint
id arxiv_https___arxiv_org_abs_2504_10258
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle XY-Cut++: Advanced Layout Ordering via Hierarchical Mask Mechanism on a Novel Benchmark
Liu, Shuai
Li, Youmeng
Wei, Jizeng
Computer Vision and Pattern Recognition
Multimedia
Document Reading Order Recovery is a fundamental task in document image understanding, playing a pivotal role in enhancing Retrieval-Augmented Generation (RAG) and serving as a critical preprocessing step for large language models (LLMs). Existing methods often struggle with complex layouts(e.g., multi-column newspapers), high-overhead interactions between cross-modal elements (visual regions and textual semantics), and a lack of robust evaluation benchmarks. We introduce XY-Cut++, an advanced layout ordering method that integrates pre-mask processing, multi-granularity segmentation, and cross-modal matching to address these challenges. Our method significantly enhances layout ordering accuracy compared to traditional XY-Cut techniques. Specifically, XY-Cut++ achieves state-of-the-art performance (98.8 BLEU overall) while maintaining simplicity and efficiency. It outperforms existing baselines by up to 24\% and demonstrates consistent accuracy across simple and complex layouts on the newly introduced DocBench-100 dataset. This advancement establishes a reliable foundation for document structure recovery, setting a new standard for layout ordering tasks and facilitating more effective RAG and LLM preprocessing.
title XY-Cut++: Advanced Layout Ordering via Hierarchical Mask Mechanism on a Novel Benchmark
topic Computer Vision and Pattern Recognition
Multimedia
url https://arxiv.org/abs/2504.10258