Beyond Isolated Dots: Benchmarking Structured Table Construction as Deep Knowledge Extraction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhong, Tianyun, Mo, Guozhao, Liu, Yanjiang, Chen, Yihan, Kong, Lingdi, Chen, Xuanang, Lu, Yaojie, Lin, Hongyu, Ye, Shiwei, Han, Xianpei, He, Ben, Sun, Le
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915586102001664
author Zhong, Tianyun
Mo, Guozhao
Liu, Yanjiang
Chen, Yihan
Kong, Lingdi
Chen, Xuanang
Lu, Yaojie
Lin, Hongyu
Ye, Shiwei
Han, Xianpei
He, Ben
Sun, Le
author_facet Zhong, Tianyun
Mo, Guozhao
Liu, Yanjiang
Chen, Yihan
Kong, Lingdi
Chen, Xuanang
Lu, Yaojie
Lin, Hongyu
Ye, Shiwei
Han, Xianpei
He, Ben
Sun, Le
contents With the emergence of large language models (LLMs), there is an expectation that LLMs can effectively extract explicit information from complex real-world documents (e.g., papers, reports). However, most LLMs generate paragraph-style answers that are chaotic, disorganized, and untraceable. To bridge this gap, we introduce the Arranged and Organized Extraction Benchmark (AOE), a new bilingual benchmark with data and documents of varying lengths designed to systematically evaluate the ability of LLMs to comprehend fragmented documents and reconstruct isolated information into one organized table. Unlike conventional text-to-table tasks, which rely on fixed schema and narrow task domains, AOE includes 11 carefully crafted tasks across three diverse domains, requiring models to generate context-specific schema tailored to varied input queries. In the experiment, we evaluated both open-source and closed-source state-of-the-art LLMs. The results show that even the most advanced models struggled significantly. The benchmark is available at https://anonymous.4open.science/r/AOE-Benchmark/.
format Preprint
id arxiv_https___arxiv_org_abs_2507_16271
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Beyond Isolated Dots: Benchmarking Structured Table Construction as Deep Knowledge Extraction
Zhong, Tianyun
Mo, Guozhao
Liu, Yanjiang
Chen, Yihan
Kong, Lingdi
Chen, Xuanang
Lu, Yaojie
Lin, Hongyu
Ye, Shiwei
Han, Xianpei
He, Ben
Sun, Le
Computation and Language
With the emergence of large language models (LLMs), there is an expectation that LLMs can effectively extract explicit information from complex real-world documents (e.g., papers, reports). However, most LLMs generate paragraph-style answers that are chaotic, disorganized, and untraceable. To bridge this gap, we introduce the Arranged and Organized Extraction Benchmark (AOE), a new bilingual benchmark with data and documents of varying lengths designed to systematically evaluate the ability of LLMs to comprehend fragmented documents and reconstruct isolated information into one organized table. Unlike conventional text-to-table tasks, which rely on fixed schema and narrow task domains, AOE includes 11 carefully crafted tasks across three diverse domains, requiring models to generate context-specific schema tailored to varied input queries. In the experiment, we evaluated both open-source and closed-source state-of-the-art LLMs. The results show that even the most advanced models struggled significantly. The benchmark is available at https://anonymous.4open.science/r/AOE-Benchmark/.
title Beyond Isolated Dots: Benchmarking Structured Table Construction as Deep Knowledge Extraction
topic Computation and Language
url https://arxiv.org/abs/2507.16271