Information Extraction From Fiscal Documents Using LLMs
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917101304807424 |
|---|---|
| author | Aggarwal, Vikram Kulkarni, Jay Mascarenhas, Aditi Narang, Aakriti Raman, Siddarth Shah, Ajay Thomas, Susan |
| author_facet | Aggarwal, Vikram Kulkarni, Jay Mascarenhas, Aditi Narang, Aakriti Raman, Siddarth Shah, Ajay Thomas, Susan |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_10659 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Information Extraction From Fiscal Documents Using LLMs Aggarwal, Vikram Kulkarni, Jay Mascarenhas, Aditi Narang, Aakriti Raman, Siddarth Shah, Ajay Thomas, Susan Computation and Language Information Retrieval 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. |
| title | Information Extraction From Fiscal Documents Using LLMs |
| topic | Computation and Language Information Retrieval |
| url | https://arxiv.org/abs/2511.10659 |