DISRetrieval: Harnessing Discourse Structure for Long Document Retrieval

2025-05-26

arXiv.org

AI Summary

In long-document question answering, existing methods treat text as flat sequences or coarse-grained chunks, ignoring discourse structure and thereby introducing semantic misalignment. This work proposes the first Rhetorical Structure Theory (RST)-based hierarchical retrieval framework, modeling documents as sentence-level tree-structured semantic graphs. Our approach features: (1) discourse-aware document organization via explicit RST tree encoding; (2) LLM-driven adaptive node representation learning; and (3) a joint evidence retrieval mechanism integrating tree-structured encoding with hierarchical attention. To our knowledge, this is the first framework to systematically incorporate linguistic discourse structure throughout the entire retrieval pipeline. Evaluated on QASPER and QuALITY benchmarks, it achieves +12.1% token-level F1 and +8.3% downstream QA accuracy over strong baselines, demonstrating the critical value of discourse structure for long-text understanding.

Technology Category

• Natural Language Processing: Question Answering
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Knowledge Representation and Reasoning: Argumentation

Application Category

• Search and Retrieval-Augmented AI: Large language models for search
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Long document understanding has become increasingly crucial in natural language processing, with retrieval-based methods emerging as a promising solution to address the context length limitations of large language models (LLMs). However, existing approaches either treat documents as flat sequences or employ arbitrary chunking strategies, failing to capture the inherent discourse structure that guides human comprehension. We present DISRetrieval, a novel hierarchical retrieval framework that leverages linguistic discourse structure to enhance long document understanding. Our approach introduces three key innovations: (1) a discourse-aware document organization framework that utilizes rhetorical structure theory (RST) to create sentence-level hierarchical representations, preserving both semantic relationships and natural document flow; (2) an LLM-enhanced node representation technique that combines discourse structure with adaptive summarization to enrich tree nodes with contextual information; and (3) a hierarchical evidence retrieval mechanism that effectively selects relevant content while maintaining discourse coherence. Through comprehensive experiments on QASPER and QuALITY datasets, DISRetrieval demonstrates substantial improvements over existing methods in both token-level retrieval metrics and downstream question answering tasks. Our ablation studies confirm that incorporating discourse structure significantly enhances retrieval effectiveness across different document lengths and query types, validating the importance of linguistically-informed document representation in long-text understanding. Our code and datasets are publicly available at github/DreamH1gh/DISRetrieval to facilitate future research.

Problem

• Capturing discourse structures for long document comprehension
• Enhancing question answering with rhetorical structure theory
• Bridging structural and semantic information using LLMs

Innovation

• Leverages rhetorical structure theory for document comprehension
• Employs LLM-enhanced node representations for semantic integration
• Uses structure-guided hierarchical retrieval for document analysis