From Surface to Semantics: Semantic Structure Parsing for Table-Centric Document Analysis

TL;DR

This paper introduces DOTABLER, a novel framework for deep semantic parsing of tables within documents, enabling advanced understanding and retrieval by uncovering semantic links between tables and their context.

Contribution

The paper presents DOTABLER, a comprehensive semantic parsing framework that integrates deep contextual understanding and domain-specific fine-tuning for improved table analysis in documents.

Findings

Achieves over 90% Precision and F1 scores on real-world PDF datasets.

Outperforms models like GPT-4o in semantic table analysis tasks.

Demonstrates effective deep parsing of tables and their contextual associations.

Abstract

Documents are core carriers of information and knowl-edge, with broad applications in finance, healthcare, and scientific research. Tables, as the main medium for structured data, encapsulate key information and are among the most critical document components. Existing studies largely focus on surface-level tasks such as layout analysis, table detection, and data extraction, lacking deep semantic parsing of tables and their contextual associations. This limits advanced tasks like cross-paragraph data interpretation and context-consistent analysis. To address this, we propose DOTABLER, a table-centric semantic document parsing framework designed to uncover deep semantic links between tables and their context. DOTABLER leverages a custom dataset and domain-specific fine-tuning of pre-trained models, integrating a complete parsing pipeline to identify context segments semantically tied to tables. Built on this semantic understanding, DOTABLER implements two core functionalities: table-centric document structure parsing and domain-specific table retrieval, delivering comprehensive table-anchored semantic analysis and precise extraction of semantically relevant tables. Evaluated on nearly 4,000 pages with over 1,000 tables from real-world PDFs, DOTABLER achieves over 90% Precision and F1 scores, demonstrating superior performance in table-context semantic analysis and deep document parsing compared to advanced models such as GPT-4o.