XY-Cut++: Advanced Layout Ordering via Hierarchical Mask Mechanism on a Novel Benchmark

TL;DR

Document Reading Order Recovery is challenged by complex document layouts and cross-modal cues between visuals and text. The paper presents XY-Cut++ a lightweight, hierarchy-aware framework that combines pre-mask processing, multi-granularity segmentation, and cross-modal matching to improve block-level reading order. It introduces the DocBench-100 benchmark and reports state-of-the-art BLEU-4 scores on both complex ($$BLEU-4$$ about $98.6$) and regular ($$BLEU-4$$ about $98.9$) layouts, with Kendall's tau near unity and strong efficiency (mean FPS around $514$). The approach yields consistent gains over classic XY-Cut and deep-learning baselines, and the work offers extensive ablations to validate each component, providing a robust foundation for RAG and LLM preprocessing in document understanding. The authors also release code to promote standardized evaluation of block-level reading-order recovery.

Abstract

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.

Figures (17)

• Figure 1: XY-Cut++ Architecture with Hierarchical Mask Mechanism and DocBench-100 Benchmark. (left) DocBench-100 dataset composition and dual evaluation protocols. (right) Algorithm workflow integrating adaptive pre-mask processing, multi-granularity segmentation, and cross-modal matching.
• Figure 2: XY-Cut Recursive Partitioning Workflow and Failure Analysis in Complex Layouts: (1) Initial document segmentation steps, (2) Connectivity assumption violations caused by cross-layout cell structures (cell 5), and (3) Error propagation through subsequent layout ordering. The correct reading order is 1, 3, 2, 4, 5, 6, 7.
• Figure 3: Challenges posed by L-shaped inputs: (1) segmentation failure due to the inability to process L-shaped structures, and (2) missegmentation caused by improper handling of L-shaped regions. The correct sequence of segmentation is ➁+➂ ➀.
• Figure 4: DocBench-100 image overview: (a) 70-page regular subset $D_r$, and (b) 30-page complex subset $D_c$. All pages provide block-level reading-order ground truth for benchmarking layout-ordering methods.
• Figure 5: End-to-End Layout Ordering for Diverse Document Layouts Framework Overview: (a) Layout Detection (PP-DocLayout pp-doclayout), (b) Pre-Mask Processing, (c) Multi-Granularity Segmentation, and (d) Cross-Modal Matching.