Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

MoDora: Tree-Based Semi-Structured Document Analysis System

Bangrui Xu, Qihang Yao, Zirui Tang, Xuanhe Zhou, Yeye He, Shihan Yu, Qianqian Xu, Bin Wang, Guoliang Li, Conghui He, Fan Wu

TL;DR

MoDora, an LLM-powered system for semi-structured document analysis, is proposed, which adopts a local-alignment aggregation strategy to convert OCR-parsed elements into layout-aware components, and designs the Component-Correlation Tree to hierarchically organize components.

Abstract

Semi-structured documents integrate diverse interleaved data elements (e.g., tables, charts, hierarchical paragraphs) arranged in various and often irregular layouts. These documents are widely observed across domains and account for a large portion of real-world data. However, existing methods struggle to support natural language question answering over these documents due to three main technical challenges: (1) The elements extracted by techniques like OCR are often fragmented and stripped of their original semantic context, making them inadequate for analysis. (2) Existing approaches lack effective representations to capture hierarchical structures within documents (e.g., associating tables with nested chapter titles) and to preserve layout-specific distinctions (e.g., differentiating sidebars from main content). (3) Answering questions often requires retrieving and aligning relevant information scattered across multiple regions or pages, such as linking a descriptive paragraph to table cells located elsewhere in the document. To address these issues, we propose MoDora, an LLM-powered system for semi-structured document analysis. First, we adopt a local-alignment aggregation strategy to convert OCR-parsed elements into layout-aware components, and conduct type-specific information extraction for components with hierarchical titles or non-text elements. Second, we design the Component-Correlation Tree (CCTree) to hierarchically organize components, explicitly modeling inter-component relations and layout distinctions through a bottom-up cascade summarization process. Finally, we propose a question-type-aware retrieval strategy that supports (1) layout-based grid partitioning for location-based retrieval and (2) LLM-guided pruning for semantic-based retrieval. Experiments show MoDora outperforms baselines by 5.97%-61.07% in accuracy. The code is at https://github.com/weAIDB/MoDora.

MoDora: Tree-Based Semi-Structured Document Analysis System

TL;DR

MoDora, an LLM-powered system for semi-structured document analysis, is proposed, which adopts a local-alignment aggregation strategy to convert OCR-parsed elements into layout-aware components, and designs the Component-Correlation Tree to hierarchically organize components.

Abstract

Semi-structured documents integrate diverse interleaved data elements (e.g., tables, charts, hierarchical paragraphs) arranged in various and often irregular layouts. These documents are widely observed across domains and account for a large portion of real-world data. However, existing methods struggle to support natural language question answering over these documents due to three main technical challenges: (1) The elements extracted by techniques like OCR are often fragmented and stripped of their original semantic context, making them inadequate for analysis. (2) Existing approaches lack effective representations to capture hierarchical structures within documents (e.g., associating tables with nested chapter titles) and to preserve layout-specific distinctions (e.g., differentiating sidebars from main content). (3) Answering questions often requires retrieving and aligning relevant information scattered across multiple regions or pages, such as linking a descriptive paragraph to table cells located elsewhere in the document. To address these issues, we propose MoDora, an LLM-powered system for semi-structured document analysis. First, we adopt a local-alignment aggregation strategy to convert OCR-parsed elements into layout-aware components, and conduct type-specific information extraction for components with hierarchical titles or non-text elements. Second, we design the Component-Correlation Tree (CCTree) to hierarchically organize components, explicitly modeling inter-component relations and layout distinctions through a bottom-up cascade summarization process. Finally, we propose a question-type-aware retrieval strategy that supports (1) layout-based grid partitioning for location-based retrieval and (2) LLM-guided pruning for semantic-based retrieval. Experiments show MoDora outperforms baselines by 5.97%-61.07% in accuracy. The code is at https://github.com/weAIDB/MoDora.
Paper Structure (24 sections, 2 equations, 9 figures, 6 tables, 3 algorithms)

This paper contains 24 sections, 2 equations, 9 figures, 6 tables, 3 algorithms.

Table of Contents

  1. Introduction
  2. Preliminaries
  3. Semi-Structured Document
  4. Semi-Structured Document Analysis
  5. Limitations of Existing Methods
  6. System Overview
  7. Document Preprocessing
  8. Local-Alignment Element Aggregation
  9. Format-Aware Hierarchy Relation Detection
  10. Template-Based Information Enrichment
  11. Tree-Based Document Modeling
  12. Component Correlation Tree
  13. Two-Stage Tree Construction
  14. Bottom-up Metadata Summarization.
  15. Tree-Based Document Analysis
  16. ...and 9 more sections

Figures (9)

  • Figure 1: Example Semi-Structured Document Analysis.
  • Figure 2: Example Semi-structured Document Layouts.
  • Figure 3: Error Rate Distribution of Existing Methods.
  • Figure 4: System Overview of MoDora.
  • Figure 5: The number distribution of summary keywords in tree nodes with different depth and subtree sizes.
  • ...and 4 more figures

Theorems & Definitions (1)

  • Definition 2.1: Semi-Structured Document Analysis