NexusSum: Hierarchical LLM Agents for Long-Form Narrative Summarization

TL;DR

NexusSum is a multi-agent LLM framework that effectively summarizes long narratives by transforming dialogue and descriptions into a unified format and employing a hierarchical pipeline, achieving state-of-the-art results.

Contribution

The paper introduces NexusSum, a novel multi-agent LLM approach with dialogue-to-description transformation and hierarchical summarization for long-form narratives, without fine-tuning.

Findings

Achieves up to 30% improvement in BERTScore across various narrative types.

Establishes new state-of-the-art in long-form narrative summarization.

Demonstrates effectiveness of multi-agent LLMs in handling complex storytelling content.

Abstract

Summarizing long-form narratives--such as books, movies, and TV scripts--requires capturing intricate plotlines, character interactions, and thematic coherence, a task that remains challenging for existing LLMs. We introduce NexusSum, a multi-agent LLM framework for narrative summarization that processes long-form text through a structured, sequential pipeline--without requiring fine-tuning. Our approach introduces two key innovations: (1) Dialogue-to-Description Transformation: A narrative-specific preprocessing method that standardizes character dialogue and descriptive text into a unified format, improving coherence. (2) Hierarchical Multi-LLM Summarization: A structured summarization pipeline that optimizes chunk processing and controls output length for accurate, high-quality summaries. Our method establishes a new state-of-the-art in narrative summarization, achieving up to a 30.0% improvement in BERTScore (F1) across books, movies, and TV scripts. These results demonstrate the effectiveness of multi-agent LLMs in handling long-form content, offering a scalable approach for structured summarization in diverse storytelling domains.