StrucSum: Graph-Structured Reasoning for Long Document Extractive Summarization with LLMs

2025/05/29 by Haohan Yuan, Yuan, Haohan, Hong, Sukhwa +2 · 1 citation
Biochemistry, Genetics and Molecular Biology · Computer Science · #Biomedical Text Mining and Ontologies #Computation and Language (cs.CL) #FOS: Computer and information sciences #Text Readability and Simplification #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2505.22950

openalex publication_date 2025/05/29 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Large language models (LLMs) have shown strong performance in zero-shot summarization, but often struggle to model document structure and identify salient information in long texts. In this work, we introduce StrucSum, a training-free prompting framework that enhances LLM reasoning through sentence-level graph structures. StrucSum injects structural signals into prompts via three targeted strategies: Neighbor-Aware Prompting (NAP) for local context, Centrality-Aware Prompting (CAP) for importance estimation, and Centrality-Guided Masking (CGM) for efficient input reduction. Experiments on ArXiv, PubMed, and Multi-News demonstrate that StrucSum consistently improves both summary quality and factual consistency over unsupervised baselines and vanilla prompting. In particular, on ArXiv, it increases FactCC and SummaC by 19.2% and 8.0% points, demonstrating stronger alignment between summaries and source content. The ablation study shows that the combination of multiple strategies does not yield clear performance gains; therefore, structure-aware prompting with graph-based information represents a promising and underexplored direction for the advancement of zero-shot extractive summarization with LLMs. Our source code is publicly available.

Citations

• QA-prompting: Improving Summarization with Large Language Models using Question-Answering
• UMB@PerAnsSumm 2025: Enhancing Perspective-Aware Summarization with Prompt Optimization and Supervised Fine-Tuning
• Discourse-Driven Evaluation: Unveiling Factual Inconsistency in Long Document Summarization
• DomainSum: A Hierarchical Benchmark for Fine-Grained Domain Shift in Abstractive Text Summarization
• FineSurE: Fine-grained Summarization Evaluation using LLMs
• Revisiting Zero-Shot Abstractive Summarization in the Era of Large Language Models from the Perspective of Position Bias
• Extractive Summarization via ChatGPT for Faithful Summary Generation
• G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment
• Benchmarking Large Language Models for News Summarization
• HEGEL: Hypergraph Transformer for Long Document Summarization
• News Summarization and Evaluation in the Era of GPT-3
• Extractive is not Faithful: An Investigation of Broad Unfaithfulness Problems in Extractive Summarization
• Large Language Models are Zero-Shot Reasoners
• HiStruct+: Improving Extractive Text Summarization with Hierarchical Structure Information
• SummaC: Re-Visiting NLI-based Models for Inconsistency Detection in Summarization
• Enhancing Extractive Text Summarization with Topic-Aware Graph Neural Networks
• Language Models are Few-Shot Learners
• Leveraging Graph to Improve Abstractive Multi-Document Summarization
• Heterogeneous Graph Neural Networks for Extractive Document Summarization
• Evaluating the Factual Consistency of Abstractive Text Summarization
• Sentence Centrality Revisited for Unsupervised Summarization
• Multi-News: a Large-Scale Multi-Document Summarization Dataset and\n Abstractive Hierarchical Model
• BERTScore: Evaluating Text Generation with BERT
• A Discourse-Aware Attention Model for Abstractive Summarization of Long Documents

Cited by

• Understanding LLM Reasoning for Abstractive Summarization
• FinTagging: Benchmarking LLMs for Extracting and Structuring Financial Information
• Curriculum-RLAIF: Curriculum Alignment with Reinforcement Learning from AI Feedback

Related

• An Evaluation of Large Language Models on Text Summarization Tasks Using Prompt Engineering Techniques
• Scaling Up Summarization: Leveraging Large Language Models for Long Text Extractive Summarization
• TriSum: Learning Summarization Ability from Large Language Models with Structured Rationale
• GoSum: Extractive Summarization of Long Documents by Reinforcement Learning and Graph Organized discourse state
• Element-aware Summarization with Large Language Models: Expert-aligned Evaluation and Chain-of-Thought Method