DTBench: A Synthetic Benchmark for Document-to-Table Extraction

TL;DR

DTBench is a synthetic benchmark designed to evaluate document-to-table extraction capabilities of language models, addressing the need for systematic, scalable, and capability-aware assessment of complex reasoning and conflict resolution tasks.

Contribution

The paper introduces DTBench, a synthetic, capability-aware benchmark for Document-to-Table extraction, created via a multi-agent synthesis workflow from ground-truth tables.

Findings

LLMs show significant performance gaps on DTBench.

Persistent challenges in reasoning and conflict resolution.

DTBench enables comprehensive evaluation of Doc2Table extraction.

Abstract

Document-to-table (Doc2Table) extraction derives structured tables from unstructured documents under a target schema, enabling reliable and verifiable SQL-based data analytics. Although large language models (LLMs) have shown promise in flexible information extraction, their ability to produce precisely structured tables remains insufficiently understood, particularly for indirect extraction that requires complex capabilities such as reasoning and conflict resolution. Existing benchmarks neither explicitly distinguish nor comprehensively cover the diverse capabilities required in Doc2Table extraction. We argue that a capability-aware benchmark is essential for systematic evaluation. However, constructing such benchmarks using human-annotated document-table pairs is costly, difficult to scale, and limited in capability coverage. To address this, we adopt a reverse Table2Doc paradigm and design a multi-agent synthesis workflow to generate documents from ground-truth tables. Based on this approach, we present DTBench, a synthetic benchmark that adopts a proposed two-level taxonomy of Doc2Table capabilities, covering 5 major categories and 13 subcategories. We evaluate several mainstream LLMs on DTBench, and demonstrate substantial performance gaps across models, as well as persistent challenges in reasoning, faithfulness, and conflict resolution. DTBench provides a comprehensive testbed for data generation and evaluation, facilitating future research on Doc2Table extraction. The benchmark is publicly available at https://github.com/ZJU-DAILY/DTBench.