On the Comprehensibility of Multi-structured Financial Documents using LLMs and Pre-processing Tools
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_ | 1866911113022537728 |
|---|---|
| author | Upadhyay, Shivani Ataey, Messiah Murtaza, Syed Shariyar Nie, Yifan Lin, Jimmy |
| author_facet | Upadhyay, Shivani Ataey, Messiah Murtaza, Syed Shariyar Nie, Yifan Lin, Jimmy |
| contents | The proliferation of complex structured data in hybrid sources, such as PDF documents and web pages, presents unique challenges for current Large Language Models (LLMs) and Multi-modal Large Language Models (MLLMs) in providing accurate answers. Despite the recent advancements of MLLMs, they still often falter when interpreting intricately structured information, such as nested tables and multi-dimensional plots, leading to hallucinations and erroneous outputs. This paper explores the capabilities of LLMs and MLLMs in understanding and answering questions from complex data structures found in PDF documents by leveraging industrial and open-source tools as part of a pre-processing pipeline. Our findings indicate that GPT-4o, a popular MLLM, achieves an accuracy of 56% on multi-structured documents when fed documents directly, and that integrating pre-processing tools raises the accuracy of LLMs to 61.3% for GPT-4o and 76% for GPT-4, and with lower overall cost. The code is publicly available at https://github.com/OGCDS/FinancialQA. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_05182 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | On the Comprehensibility of Multi-structured Financial Documents using LLMs and Pre-processing Tools Upadhyay, Shivani Ataey, Messiah Murtaza, Syed Shariyar Nie, Yifan Lin, Jimmy Information Retrieval The proliferation of complex structured data in hybrid sources, such as PDF documents and web pages, presents unique challenges for current Large Language Models (LLMs) and Multi-modal Large Language Models (MLLMs) in providing accurate answers. Despite the recent advancements of MLLMs, they still often falter when interpreting intricately structured information, such as nested tables and multi-dimensional plots, leading to hallucinations and erroneous outputs. This paper explores the capabilities of LLMs and MLLMs in understanding and answering questions from complex data structures found in PDF documents by leveraging industrial and open-source tools as part of a pre-processing pipeline. Our findings indicate that GPT-4o, a popular MLLM, achieves an accuracy of 56% on multi-structured documents when fed documents directly, and that integrating pre-processing tools raises the accuracy of LLMs to 61.3% for GPT-4o and 76% for GPT-4, and with lower overall cost. The code is publicly available at https://github.com/OGCDS/FinancialQA. |
| title | On the Comprehensibility of Multi-structured Financial Documents using LLMs and Pre-processing Tools |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2506.05182 |