@inproceedings{zhou-etal-2025-reso,
    title = "{R}e{S}o: A Reward-driven Self-organizing {LLM}-based Multi-Agent System for Reasoning Tasks",
    author = "Zhou, Heng  and
      Geng, Hejia  and
      Xue, Xiangyuan  and
      Kang, Li  and
      Qin, Yiran  and
      Wang, Zhiyong  and
      Yin, Zhenfei  and
      Bai, Lei",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.emnlp-main.808/",
    pages = "15990--16009",
    ISBN = "979-8-89176-332-6",
    abstract = "Multi-agent systems have emerged as a promising approach for enhancing the reasoning capabilities of large language models in complex problem-solving. However, current MAS frameworks are limited by poor flexibility and scalability, with underdeveloped optimization strategies. To address these challenges, we propose ReSo, which integrates task graph generation with a reward-driven two-stage agent selection process. The core of ReSo is the proposed Collaborative Reward Model, which can provide fine-grained reward signals for MAS cooperation for optimization. We also introduce an automated data synthesis framework for generating MAS benchmarks, without human annotations. Experimentally, ReSo matches or outperforms existing methods. ReSo achieves $\textbf{33.7\%}$ and $\textbf{32.3\%}$ accuracy on Math-MAS and SciBench-MAS SciBench, while other methods completely fail. The code and data are available at [Reso](https://github.com/hengzzzhou/ReSo)."
}