---
title: "MA-RWG: A Multi-Agent Framework for Thematically Structuring and Generation of Related Work"
authors:
  - "Zhuang Liu"
  - "Jian Liu"
  - "Chun Kang"
  - "Chenbin Zhang"
  - "Rui Li"
  - "Fanhu Zeng"
  - "Yong Dai"
  - "Lei Sha"
doi: "10.24963/ijcai.2026/656"
url: "https://ijcai.org/proceedings/2026/656"
pdf: "https://ijcai.org/proceedings/2026/0656.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "5892-5900"
track: "Main Track"
keywords:
  - "Natural Language Processing: Information retrieval and text mining"
  - "Agent-based and Multi-agent Systems: Applications"
  - "Natural Language Processing: Summarization"
  - "Natural Language Processing: Language generation"
---

# MA-RWG: A Multi-Agent Framework for Thematically Structuring and Generation of Related Work

Zhuang Liu, Jian Liu, Chun Kang, Chenbin Zhang, Rui Li, Fanhu Zeng, Yong Dai, Lei Sha

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 5892-5900. DOI: [10.24963/ijcai.2026/656](https://doi.org/10.24963/ijcai.2026/656)

## Abstract

AI-driven survey generation has advanced rapidly, yet related work generation (RWG) remains relatively underexplored. Unlike surveys that provide broad literature overviews, RWG synthesizes prior studies for a single focal paper, requiring contextual fit, cross-paper comparison, and accurate attribution. To address this gap, we propose MA-RWG, a fully automated multi-agent framework that generates polished related work sections from only a title and abstract. MA-RWG first retrieves high-quality candidate papers through semantic retrieval, optionally enhanced with a diversity-aware term. It then coordinates four specialized agents for summarization, organization, integration, and fact checking, enabling DAG-based taxonomy construction, feedback-guided refinement, and dual-model verification. For evaluation, we introduce a dedicated benchmark for paper-specific related work generation, covering generation quality, citation quality, and claim-level semantic similarity. Experimental results show that MA-RWG outperforms RAG-based baselines and survey-oriented agentic methods on the RWG task. Further ablation and cross-domain experiments demonstrate the soundness and robustness of the proposed framework.