paper-with-me

홈 › Papers

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

2024-02-19 · Naihao Deng, Zhenjie Sun, Ruiqi He, Aman Sikka, Yulong Chen, Lin Ma, Yue Zhang, Rada Mihalcea

In this paper, we investigate the effectiveness of various LLMs in interpreting tabular data through different prompting strategies and data formats. Our analyses extend across six benchmarks for table-related tasks such as question-answering and fact-checking. We introduce for the first time the assessment of LLMs' performance on image-based table representations. Specifically, we compare five text-based and three image-based table representations, demonstrating the role of representation and prompting on LLM performance. Our study provides insights into the effective use of LLMs on table-related tasks.

📄 PDF Abstract BibTeX arXiv:2402.12424

Code (0)

등록된 구현이 없습니다.

Tasks

Fact CheckingQuestion Answering

Similar Papers 제목 키워드 기반

Logical Inference for Counting on Semi-structured Tables

2022-04-16 · ACL 2022 5 · Tomoya Kurosawa, Hitomi Yanaka

Recently, the Natural Language Inference (NLI) task has been studied for semi-structured tables that do not have a strict format. Although neural approaches have achieved high performance in various types of NLI, includi…

Natural Language Inference

A Table Is Worth 64 Tokens: Pixel-level Compression for Multi-Table Document Question Answering

2026-08-27 · Iñigo Alonso, Mirella Lapata arxiv

Answering questions over real-world documents requires processing long inputs that interleave text with tables. Optical context compression, which represents context as images, promises to reduce token cost, but its effe…

Question Answering

TeleTables: A Benchmark for Large Language Models in Telecom Table Interpretation

2025-12-05 · Anas Ezzakri, Nicola Piovesan, Mohamed Sana, Antonio De Domenico 외 arxiv

Large Language Models (LLMs) are increasingly applied to telecom engineering tasks, yet perform poorly on 3GPP specifications. These standards encode much of their technical information in complex tables, but LLM knowled…

TransientTables: Evaluating LLMs' Reasoning on Temporally Evolving Semi-structured Tables

2025-04-02 · Abhilash Shankarampeta, Harsh Mahajan, Tushar Kataria, Dan Roth 외

Humans continuously make new discoveries, and understanding temporal sequence of events leading to these breakthroughs is essential for advancing science and society. This ability to reason over time allows us to identif…

Question GenerationQuestion-Generation

DocMath-Eval: Evaluating Math Reasoning Capabilities of LLMs in Understanding Long and Specialized Documents

2023-11-16 · Yilun Zhao, Yitao Long, Hongjun Liu, Ryo Kamoi 외

Recent LLMs have demonstrated remarkable performance in solving exam-like math word problems. However, the degree to which these numerical reasoning skills are effective in real-world scenarios, particularly in expert do…

Math