Information Extraction From Fiscal Documents Using LLMs

TL;DR

This paper introduces a novel LLM-based method for extracting and validating structured data from complex, multi-page fiscal documents, demonstrating high accuracy and robustness in hierarchical data processing.

Contribution

It presents a multi-stage pipeline leveraging LLMs, domain knowledge, and hierarchical validation to extract structured fiscal data from lengthy government documents.

Findings

High accuracy in data extraction from multi-page fiscal documents

Effective validation using hierarchical relationships within fiscal tables

Scalable approach applicable to developing country contexts

Abstract

Large Language Models (LLMs) have demonstrated remarkable capabilities in text comprehension, but their ability to process complex, hierarchical tabular data remains underexplored. We present a novel approach to extracting structured data from multi-page government fiscal documents using LLM-based techniques. Applied to annual fiscal documents from the State of Karnataka in India (200+ pages), our method achieves high accuracy through a multi-stage pipeline that leverages domain knowledge, sequential context, and algorithmic validation. A large challenge with traditional OCR methods is the inability to verify the accurate extraction of numbers. When applied to fiscal data, the inherent structure of fiscal tables, with totals at each level of the hierarchy, allows for robust internal validation of the extracted data. We use these hierarchical relationships to create multi-level validation checks. We demonstrate that LLMs can read tables and also process document-specific structural hierarchies, offering a scalable process for converting PDF-based fiscal disclosures into research-ready databases. Our implementation shows promise for broader applications across developing country contexts.