DeepRead: Document Structure-Aware Reasoning to Enhance Agentic Search

TL;DR

DeepRead introduces a structure-aware reasoning agent that leverages document hierarchy and sequential logic to improve multi-turn evidence gathering in large language models, outperforming traditional unstructured retrieval methods.

Contribution

It presents DeepRead, a novel document reasoning framework that incorporates native document structure into LLM-based agentic search, enabling more human-like, accurate document comprehension.

Findings

DeepRead outperforms baseline methods by 10.3% on four benchmarks.

The system effectively mimics human locate-then-read strategies.

Structural awareness enhances reasoning accuracy in document retrieval.

Abstract

With the rapid advancement of tool-use capabilities in Large Language Models (LLMs), Retrieval-Augmented Generation (RAG) is shifting from static, one-shot retrieval toward autonomous, multi-turn evidence acquisition. However, existing agentic search frameworks typically treat long documents as flat collections of unstructured chunks, disregarding the native hierarchical organization and sequential logic essential for human comprehension. To bridge this gap, we introduce \textbf{DeepRead}, a structure-aware document reasoning agent designed to operationalize document-native structural priors into actionable reasoning capabilities. Leveraging the structural fidelity of modern OCR, DeepRead constructs a paragraph-level, coordinate-based navigation system and equips the LLM with two synergistic tools: \textsf{Retrieve} for scanning-aware localization, and \textsf{ReadSection} for contiguous, order-preserving reading within specific hierarchical scopes. This design elicits a human-like ``locate-then-read'' reasoning paradigm, effectively mitigating the context fragmentation inherent in traditional retrieval methods. Extensive evaluations across four benchmarks spanning diverse document types demonstrate that DeepRead outperforms Search-o1-style agentic search baselines by an average of 10.3\%. Fine-grained behavioral analysis further confirms that DeepRead autonomously adopts human-aligned reading strategies, validating the critical role of structural awareness in achieving precise document reasoning. Our code is available at https://github.com/Zhanli-Li/DeepRead.