Table Understanding and (Multimodal) LLMs: A Cross-Domain Case Study on Scientific vs. Non-Scientific Data

Ekaterina Borisova, Fabio Barth, Nils Feldhus, Raia Abu Ahmad, Malte Ostendorff +3 more
6/30/2025
cs.CL

No official code found. Best implementation: esborisova/TableEval-Study (3 stars, MIT, last commit 2025-07-25).

Abstract

Tables are among the most widely used tools for representing structured data in research, business, medicine, and education. Although LLMs demonstrate strong performance in downstream tasks, their efficiency in processing tabular data remains underexplored. In this paper, we investigate the effectiveness of both text-based and multimodal LLMs on table understanding tasks through a cross-domain and cross-modality evaluation. Specifically, we compare their performance on tables from scientific vs. non-scientific contexts and examine their robustness on tables represented as images vs. text. Additionally, we conduct an interpretability analysis to measure context usage and input relevance. We also introduce the TableEval benchmark, comprising 3017 tables from scholarly publications, Wikipedia, and financial reports, where each table is provided in five different formats: Image, Dictionary, HTML, XML, and LaTeX. Our findings indicate that while LLMs maintain robustness across table modalities, they face significant challenges when processing scientific tables.

View on arXivView PDF

Code Implementations(1)

esborisova/TableEval-StudyCommunity62%
MIT

Table Understanding and (Multimodal) LLMs: A Cross-Domain Case Study on Scientific vs. Non-Scientific Data

31PythonAug 27, 20241 years agoMIT
benchmarkingmultimodal-large-language-modelstable-understanding

Similar Papers

Tables as Texts or Images: Evaluating the Table Reasoning Ability of LLMs and MLLMs

88% match2024

Tabular Data Understanding with LLMs: A Survey of Recent Advances and Challenges

86% match2025

Does Table Source Matter? Benchmarking and Improving Multimodal Scientific Table Understanding and Reasoning

85% match1 repo42025

Format Matters: The Robustness of Multimodal LLMs in Reviewing Evidence from Tables and Charts

85% match2025

Benchmarking Multimodal LLMs on Recognition and Understanding over Chemical Tables

84% match2025

Multimodal Table Understanding

83% match1 repo2272024

Cite this paper

@article{borisova2025table,
  title  = {Table Understanding and (Multimodal) LLMs: A Cross-Domain Case Study on Scientific vs. Non-Scientific Data},
  author = {Ekaterina Borisova and Fabio Barth and Nils Feldhus and Raia Abu Ahmad and Malte Ostendorff and Pedro Ortiz Suarez and Georg Rehm and Sebastian Möller},
  year   = {2025},
  eprint = {2507.00152},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2507.00152v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2507.00152.svg)](https://codeofpaper.com/paper/2507.00152)

Discussion