Robust Hierarchical Structures for Agentic Document Analysis

原文：Robust Hierarchical Structures for Agentic Document Analysis

作者：Ruiying Ma, Yiming Lin, Aditya G. Parameswaran

来源：arXiv cs.DB（数据库）

正文

Computer Science > Databases

arXiv:2609.33322v1 (cs)

[Submitted on 27 Sep 2026]

Title:Robust Hierarchical Structures for Agentic Document Analysis

Authors:Ruiying Ma, Yiming Lin, Aditya G. Parameswaran

View a PDF of the paper titled Robust Hierarchical Structures for Agentic Document Analysis, by Ruiying Ma and 2 other authors

View PDF

HTML (experimental)

Abstract:Large Language Models (LLMs) enable us to better understand text documents, including PDFs and Word documents. However, LLMs, as well as more modern LLM agents, i.e., those with tool-calling abilities, typically treat such documents as plain text, ignoring the fact that they are often organized hierarchically into sections and subsections. Extracting this structure, while difficult, can improve efficiency and effectiveness for agents (and humans)—since only sections relevant to a given task need to be processed. Unfortunately, prior work on structure extraction provides no formal guarantees on how well the inferred structure matches the true one. Instead, we target a robust and compact variant that is feasible to infer and useful in practice. Robustness ensures that the text under each subsection header is a superset of the text under the same header in the true structure. Compactness seeks to minimize this superset, reducing agentic cost (or human cognitive load). We propose SHED, a two-stage workflow for inferring a robust and compact structure. The first stage is pluggable with an infinite family of approaches, each guaranteeing robustness for a specific document class. We theoretically characterize the document space using these classes and their hierarchical relationships. Empirically, SHED improves F-1 scores (measuring the robustness–compactness trade-off) by 13%–68% over non-LLM baselines and 9%–15% over expensive LLM-based approaches. Finally, we show how SHED-inferred structures are valuable for agentic document analysis: agents using SHED outperform baselines, achieving 3%–23% higher accuracy while being up to 10x cheaper.

Comments:

To appear in Proceedings of the ACM on Management of Data (SIGMOD 2027). 25 pages

Subjects:

Databases (cs.DB); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Information Retrieval (cs.IR)

Cite as:

arXiv:2609.33322 [cs.DB]

(or

arXiv:2609.33322v1 [cs.DB] for this version)

https://doi.org/10.48550/arXiv.2609.33322

Focus to learn more

arXiv-issued DOI via DataCite (pending registration)

主题

数据库 · 人工智能 · 自然语言处理 · 信息检索

由「前沿雷达」于 2026-09-27 采集。正文取自原文页面，已保留出处链接。标题与正文版权归原作者所有。