A Training-Free Mixture-of-Agents Framework for Multi-Document Summarization using LLMs and Knowledge Graphs
2026-06-02 • Computation and Language
Computation and LanguageArtificial Intelligence
AI summaryⓘ
The authors created a new method to summarize information from multiple documents without needing extra training. They use different specialized 'agents' that each handle parts of the task, like picking important info, adding knowledge-based details, and improving the summary step-by-step. These agents work together using large language models and knowledge graphs to produce reliable summaries. Their approach works well for different languages and topics, based on tests with several datasets.
Multi-Document SummarizationLarge Language ModelsKnowledge GraphsExtractive SummarizationAbstractive SummarizationIterative RefinementModular DesignZero-shot LearningCross-lingual NLP
Authors
Cuong Vuong Tuan, Trang Mai Xuan, Tien-Cuong Nguyen, Vu-Duc Ngo, Thien Van Luong
Abstract
Multi-Document Summarization (MDS) plays a critical role in distilling essential information from collections of textual data. Existing approaches often struggle to capture complex inter-document relationships, rely heavily on large amounts of labeled data for supervised training, or exhibit limited generalization across domains and languages. To address these limitations, we present a training-free mixture-of-agents framework for MDS that leverages the complementary strengths of large language models (LLMs) and knowledge graphs. Our approach decomposes summarization into specialized agent tasks: extractive selection, knowledge-aware abstraction, and iterative refinement, each operating without task-specific fine-tuning. We unify their outputs using a multi-perspective consistency mechanism guided by LLMs. Experiments across four datasets in English and Vietnamese demonstrate state-of-the-art or competitive performance, validating the effectiveness and adaptability of our modular design.