Elamur Research⌕
☰
home / papers / 2026 / TopoDIM: One-shot Topology Generation of Diverse Interaction Modes for Multi-Agent Systems

TopoDIM: One-shot Topology Generation of Diverse Interaction Modes for Multi-Agent Systems

Paper · 2026 · Rui Sun, Jie Ding, Chenghua Gong et al.

Summary

Communication topologies in LLM-based multi-agent systems (MAS) usually depend on spatio-temporal designs that mix intra- and inter-round dialogue. Running many rounds in sequence costs a lot of tokens and time. TopoDIM replaces this with a single-step approach: it builds a heterogeneous communication graph in one inference pass. Each edge in the graph carries one of three interaction modes: conditioned, feedback, or debate. To produce the graph, a relational graph convolutional encoder reads a prior knowledge graph initialised by an LLM. An autoregressive decoder then assigns an edge type, or no edge, to every agent pair, with a mask that keeps the graph acyclic and an adaptive sparsification step. The global policy is trained with reinforcement learning and then distilled into small local networks on each agent. This lets agents form their connections without a central coordinator, which helps privacy and adaptability. Compared with state-of-the-art methods, TopoDIM uses fewer tokens and slightly improves accuracy. It also copes well when agents run on different LLMs, which makes it relevant for building cheaper collaborative LLM systems.

Key points

  • Compared with the strongest baselines, total token consumption drops by 46.41% and average task performance rises by 1.50%.
  • There are three edge types:
    • Conditioned: an agent uses another agent's output as input.
    • Feedback: one agent critiques another, which then revises its answer.
    • Debate: a two-round challenge, after which the challenger answers.
  • Training uses a policy gradient with entropy regularization and a reward that combines task performance and structural diversity. Teacher-to-student distillation minimises the KL divergence between the global and local policies.
  • The work is aimed at both homogeneous and heterogeneous agent pools. The authors report that mixing LLMs of different strength helps avoid a "bucket effect", where the weakest agent limits the whole system.
  • The paper was published in Findings of ACL 2026 (pp. 4252–4269) by authors from USTC and ECNU. Code is available at github.com/Sundiasy/TopoDIM.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓University of Science and Technology of ChinaHefei · China145 works · 114,409 citations🎓East China Normal UniversityShanghai · China69 works · 2,490 citations

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 worksArts and Humanities · 17,450 worksLanguage and Linguistics · 17,296 worksComputer Networks and Communications · 100 works

Cite

Cite as Rui Sun, Jie Ding, Chenghua Gong, Tianjun Gu, Yihang Jiang, Juyuan Zhang, Liming Pan and Linyuan Lü. 2026. TopoDIM: One-shot Topology Generation of Diverse Interaction Modes for Multi-Agent Systems. Findings of the Association for Computational Linguistics: ACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.findings-acl.207

@inproceedings{sun2026topodim,
  title = {{TopoDIM: One-shot Topology Generation of Diverse Interaction Modes for Multi-Agent Systems}},
  author = {Rui Sun, Jie Ding, Chenghua Gong, Tianjun Gu, Yihang Jiang, Juyuan Zhang, Liming Pan and Linyuan Lü},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.207},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.207}
}

References

59 references, 23 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Sandhini Agarwal, Lama Ahmad, Jason Ai, Sam Altman, Andy Applebaum, Edwin Arbus, Rahul K Arora, Yu Bai, Bowen Baker, Haiming Bao, and 1 others. 2025. gpt-oss-120b & gpt-oss-20b model card. arXiv preprint arXiv:2508.10925.
  2. Jinchi Chen, Jie Feng, Weiguo Gao, and Ke Wei. 2024. Decentralized natural policy gradient with variance reduction for collaborative multi-agent reinforcement learning. Journal of Machine Learning Research, 25(172):1–49. find
  3. Weize Chen, Yusheng Su, Jingwei Zuo, Cheng Yang, Chenfei Yuan, Chi-Min Chan, Heyang Yu, Yaxi Lu, Yi-Hsin Hung, Chen Qian, and 1 others. 2023. Agentverse: Facilitating multi-agent collaboration and exploring emergent behaviors. In The Twelfth International Conference on Learning Representations.
  4. Weize Chen, Jiarui Yuan, Chen Qian, Cheng Yang, Zhiyuan Liu, and Maosong Sun. 2025. Optima: Optimizing effectiveness and efficiency for llm-based multi-agent system. In Findings of the Association for Computational Linguistics: ACL 2025, pages 11534–11557. find
  5. Hyeong Kyu Choi, Xiaojin Zhu, and Yixuan Li. 2025. Debate or vote: Which yields better decisions in multi-agent large language models? arXiv preprint arXiv:2508.17536. — [arXiv:2508.17536](https://arxiv.org/abs/2508.17536) find
  6. Karl Cobbe, Vineet Kosaraju, Mohammad Bavarian, Mark Chen, Heewoo Jun, Lukasz Kaiser, Matthias Plappert, Jerry Tworek, Jacob Hilton, Reiichiro Nakano, and 1 others. 2021. Training verifiers to solve math word problems. arXiv preprint arXiv:2110.14168.
  7. Yilun Du, Shuang Li, Antonio Torralba, Joshua B Tenenbaum, and Igor Mordatch. 2023. Improving factuality and reasoning in language models through multiagent debate. In Forty-first International Conference on Machine Learning.
  8. Tianyu Fu, Zihan Min, Hanling Zhang, Jichao Yan, Guohao Dai, Wanli Ouyang, and Yu Wang. 2025. Cache-to-cache: Direct semantic communication between large language models. arXiv preprint arXiv:2510.03215. — [arXiv:2510.03215](https://arxiv.org/abs/2510.03215) find
  9. Ali Essam Ghareeb, Benjamin Chang, Ludovico Mitchener, Angela Yiu, Caralyn J Szostkiewicz, Jon M Laurent, Muhammed T Razzak, Andrew D White, Michaela M Hinks, and Samuel G Rodriques. 2025. Robin: A multi-agent system for automating scientific discovery. arXiv preprint arXiv:2505.13400. find
  10. Eric Hanchen Jiang, Guancheng Wan, Sophia Yin, Mengting Li, Yuchen Wu, Xiao Liang, Xinfeng Li, Yizhou Sun, Wei Wang, Kai-Wei Chang, and 1 others. 2025. Dynamic generation of multi-llm agents communication topologies with graph diffusion models. arXiv preprint arXiv:2510.07799. find
  11. Chenghua Gong and Xiang Li. 2025. Agents with foundation models: advance and vision. Frontiers of Computer Science, 19(4):194330. find
  12. Bin Lei, Yi Zhang, Shan Zuo, Ali Payani, and Caiwen Ding. 2024. Macm: Utilizing a multi-agent system for condition mining in solving complex mathematical problems. Advances in Neural Information Processing Systems, 37:53418–53437. find
  13. Junda He, Christoph Treude, and David Lo. 2025. Llm-based multi-agent systems for software engineering: Literature review, vision, and the road ahead. ACM Transactions on Software Engineering and Methodology, 34(5):1–30. find
  14. Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, and Jacob Steinhardt. 2020. Measuring massive multitask language understanding. arXiv preprint arXiv:2009.03300.
  15. Samuel Holt, Max Ruiz Luyten, and Mihaela van der Schaar. 2025. L2mac: Large language model automatic computer for extensive code generation. arXiv preprint arXiv:2310.02003. — [arXiv:2310.02003](https://arxiv.org/abs/2310.02003) find
  16. Sirui Hong, Mingchen Zhuge, Jonathan Chen, Xiawu Zheng, Yuheng Cheng, Jinlin Wang, Ceyao Zhang, Zili Wang, Steven Ka Shing Yau, Zijuan Lin, and 1 others. 2023. Metagpt: Meta programming for a multi-agent collaborative framework. In The Twelfth International Conference on Learning Representations.
  17. Jinwei Hu, Yi Dong, Youcheng Sun, and Xiaowei Huang. 2026a. Tapas are free! training-free adaptation of programmatic agents via llm-guided program synthesis in dynamic environments. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 40, pages 29477–29485. find
  18. Jinwei Hu, Xinmiao Huang, Youcheng Sun, Yi Dong, and Xiaowei Huang. 2026b. Lying with truths: Open-channel multi-agent collusion for belief manipulation via generative montage. arXiv preprint arXiv:2601.01685. find
  19. Junan Hu, Shudan Guo, Wenqi Liu, Jianhua Yin, and Yinwei Wei. 2026c. Context-agent: Dynamic discourse trees for non-linear dialogue. arXiv preprint arXiv:2604.05552. find
  20. Guohao Li, Hasan Hammoud, Hani Itani, Dmitrii Khizbullin, and Bernard Ghanem. 2023. Camel: Communicative agents for" mind" exploration of large language model society. Advances in Neural Information Processing Systems, 36:51991–52008.
  21. Shiyuan Li, Yixin Liu, Qingsong Wen, Chengqi Zhang, and Shirui Pan. 2025. Assemble your crew: Automatic multi-agent communication topology design via autoregressive graph generation. arXiv preprint arXiv:2507.18224. — [arXiv:2507.18224](https://arxiv.org/abs/2507.18224) find
  22. Tian Liang, Zhiwei He, Wenxiang Jiao, Xing Wang, Yan Wang, Rui Wang, Yujiu Yang, Shuming Shi, and Zhaopeng Tu. 2023. Encouraging divergent thinking in large language models through multi-agent debate. arXiv preprint arXiv:2305.19118.
  23. Aixin Liu, Aoxue Mei, Bangcai Lin, Bing Xue, Bingxuan Wang, Bingzheng Xu, Bochao Wu, Bowei Zhang, Chaofan Lin, Chen Dong, and 1 others. 2025. Deepseek-v3. 2: Pushing the frontier of open large language models. arXiv preprint arXiv:2512.02556.
  24. Shiyu Liu, Yongjing Yin, Jianhao Yan, Yunbo Tang, Qinggang Zhang, Bei Li, Xin Chen, Jingang Wang, Xunliang Cai, and Jinsong Su. 2026. Bapo: Boundary-aware policy optimization for reliable agentic search. arXiv preprint arXiv:2601.11037. find
  25. Zijun Liu, Yanzhe Zhang, Peng Li, Yang Liu, and Diyi Yang. 2024. A dynamic llm-powered agent network for task-oriented agent collaboration. In First Conference on Language Modeling.
  26. Yunbo Long, Yuhan Liu, and Liming Xu. 2026. Emomas: Emotion-aware multi-agent system for high-stakes edge-deployable negotiation with bayesian orchestration. arXiv preprint arXiv:2604.07003.
  27. Md Ashraful Islam, Mohammed Eunus Ali, and Md Rizwan Parvez. 2024. Mapcoder: Multi-agent code generation for competitive problem solving. arXiv preprint arXiv:2405.11403.
  28. Shichao Ma, Yunhe Guo, Jiahao Su, Qihe Huang, Zhengyang Zhou, and Yang Wang. 2026a. Talk2image: A multi-agent system for multi-turn image generation and editing. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 40, pages 32437–32445. find
  29. Naman Jain, King Han, Alex Gu, Wen-Ding Li, Fanjia Yan, Tianjun Zhang, Sida Wang, Armando Solar-Lezama, Koushik Sen, and Ion Stoica. 2024. Livecodebench: Holistic and contamination free evaluation of large language models for code. arXiv preprint arXiv:2403.07974.
  30. Shichao Ma, Zhiyuan Ma, Ming Yang, Xiaofan Li, Xing Wu, Jintao Du, Yu Cheng, Weiqiang Wang, Qiliang Liu, Zhengyang Zhou, and 1 others. 2026b. Tspo: Breaking the double homogenization dilemma in multi-turn search policy optimization. arXiv preprint arXiv:2601.22776.
  31. Mathematical Association of America. 1983–2025. American invitational mathematics examination (aime). Accessed via Art of Problem Solving or [Insert Dataset Name]. find
  32. Ollama. 2023. Ollama: Large language model runner. Software available at https://github.com/ollama/ollama. ↗
  33. OpenAI. 2025. Gpt-5 system card. https://openai.com/index/gpt-5-system-card/. Accessed: 2025-08-13. ↗
  34. Chen Qian, Wei Liu, Hongzhang Liu, Nuo Chen, Yufan Dang, Jiahao Li, Cheng Yang, Weize Chen, Yusheng Su, Xin Cong, and 1 others. 2024a. Chatdev: Communicative agents for software development. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Pa…
  35. Chen Qian, Zihao Xie, Yifei Wang, Wei Liu, Kunlun Zhu, Hanchen Xia, Yufan Dang, Zhuoyun Du, Weize Chen, Cheng Yang, and 1 others. 2024b. Scaling large language model-based multi-agent collaboration. arXiv preprint arXiv:2406.07155.
  36. Subhro Roy and Dan Roth. 2016. Solving general arithmetic word problems. arXiv preprint arXiv:1608.01413.
  37. Marlene Scardamalia and Carl Bereiter. 2006. Knowledge building. The Cambridge. find
  38. Michael Schlichtkrull, Thomas N Kipf, Peter Bloem, Rianne Van Den Berg, Ivan Titov, and Max Welling. 2018. Modeling relational data with graph convolutional networks. In European semantic web conference, pages 593–607. Springer. find
  39. Richard S Sutton, David McAllester, Satinder Singh, and Yishay Mansour. 1999. Policy gradient methods for reinforcement learning with function approximation. Advances in neural information processing systems, 12. find
  40. Wenhui Wang, Furu Wei, Li Dong, Hangbo Bao, Nan Yang, and Ming Zhou. 2020. Minilm: Deep self-attention distillation for task-agnostic compression of pre-trained transformers. Advances in neural information processing systems, 33:5776–5788. find
  41. Yu Wang and Xi Chen. 2025. Mirix: Multi-agent memory system for llm-based agents. arXiv preprint arXiv:2507.07957. — [arXiv:2507.07957](https://arxiv.org/abs/2507.07957) find
  42. Yubo Wang, Xueguang Ma, Ge Zhang, Yuansheng Ni, Abhranil Chandra, Shiguang Guo, Weiming Ren, Aaran Arulraj, Xuan He, Ziyan Jiang, and 1 others. 2024b. Mmlu-pro: A more robust and challenging multi-task language understanding benchmark. Advances in Neural Information Processing Systems, 37:95266–9529…
  43. Zhexuan Wang, Yutong Wang, Xuebo Liu, Liang Ding, Miao Zhang, Jie Liu, and Min Zhang. 2025b. Agentdropout: Dynamic agent elimination for token-efficient and high-performance llm-based multi-agent collaboration. arXiv preprint arXiv:2503.18891.
  44. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, and 1 others. 2022. Chain-of-thought prompting elicits reasoning in large language models. Advances in neural information processing systems, 35:24824– 24837.
  45. Ronald J Williams. 1992. Simple statistical gradient-following algorithms for connectionist reinforcement learning. Machine learning, 8(3):229–256. find
  46. Qingyun Wu, Gagan Bansal, Jieyu Zhang, Yiran Wu, Shaokun Zhang, Erkang Zhu, Beibin Li, Li Jiang, Xiaoyun Zhang, and Chi Wang. 2023. Autogen: Enabling next-gen llm applications via multi-agent conversation framework. arXiv preprint arXiv:2308.08155, 3(4).
  47. Xingjian Wu, Xvyuan Liu, Junkai Lu, Siyuan Wang, Xiangfei Qiu, Yang Shu, Jilin Hu, Chenjuan Guo, and Bin Yang. 2026. St-evo: Towards generative spatiotemporal evolution of multi-agent communication topologies. arXiv preprint arXiv:2602.14681. — [arXiv:2602.14681](https://arxiv.org/abs/2602.14681) find
  48. Gemma Team, Aishwarya Kamath, Johan Ferret, Shreya Pathak, Nino Vieillard, Ramona Merhej, Sarah Perrin, Tatiana Matejovicova, Alexandre Ramé, Morgane Rivière, and 1 others. 2025. Gemma 3 technical report. arXiv preprint arXiv:2503.19786. find
  49. Xiangyuan Xue, Yifan Zhou, Guibin Zhang, Zaibin Zhang, Yijiang Li, Chen Zhang, Zhenfei Yin, Philip Torr, Wanli Ouyang, and Lei Bai. 2025. Comas: Co-evolving multi-agent systems via interaction rewards. arXiv preprint arXiv:2510.08529. — [arXiv:2510.08529](https://arxiv.org/abs/2510.08529) find
  50. Fengyi Wang, Guanghui Zhu, Chunfeng Yuan, and Yihua Huang. 2024a. Llm-enhanced cascaded multi-level learning on temporal heterogeneous graphs. In Proceedings of the 47th International ACM SIGIR Conference on Research and Development in Information Retrieval, pages 512–521. find
  51. Yingxuan Yang, Huacan Chai, Shuai Shao, Yuanyi Song, Siyuan Qi, Renting Rui, and Weinan Zhang. 2025. Agentnet: Decentralized evolutionary coordination for llm-based multi-agent systems. arXiv preprint arXiv:2504.00587. — [arXiv:2504.00587](https://arxiv.org/abs/2504.00587) find
  52. Song Wang, Zhen Tan, Zihan Chen, Shuang Zhou, Tianlong Chen, and Jundong Li. 2025a. Anymac: Cascading flexible multi-agent collaboration via next-agent prediction. arXiv preprint arXiv:2506.17784.
  53. Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik R Narasimhan, and Yuan Cao. 2022. React: Synergizing reasoning and acting in language models. In The eleventh international conference on learning representations.
  54. Rui Ye, Xiangrui Liu, Qimin Wu, Xianghe Pang, Zhenfei Yin, Lei Bai, and Siheng Chen. 2025. X-mas: Towards building multi-agent systems with heterogeneous llms. arXiv preprint arXiv:2505.16997. — [arXiv:2505.16997](https://arxiv.org/abs/2505.16997) find
  55. Yuting Zeng, Weizhe Huang, Lei Jiang, Tongxuan Liu, Xitai Jin, Chen Tianying Tiana, Jing Li, and Xiaohua Xu. 2025. S 2 -mad: Breaking the token barrier to enhance multi-agent debate efficiency. arXiv preprint arXiv:2502.04790. find
  56. Guibin Zhang, Muxin Fu, Guancheng Wan, Miao Yu, Kun Wang, and Shuicheng Yan. 2025. G-memory: Tracing hierarchical memory for multi-agent systems. arXiv preprint arXiv:2506.07398. — [arXiv:2506.07398](https://arxiv.org/abs/2506.07398) find
  57. Guibin Zhang, Yanwei Yue, Zhixun Li, Sukwon Yun, Guancheng Wan, Kun Wang, Dawei Cheng, Jeffrey Xu Yu, and Tianlong Chen. 2024a. Cut the crap: An economical communication pipeline for llm-based multi-agent systems. arXiv preprint arXiv:2410.02506. — [arXiv:2410.02506](https://arxiv.org/abs/2410.02506) find
  58. Guibin Zhang, Yanwei Yue, Xiangguo Sun, Guancheng Wan, Miao Yu, Junfeng Fang, Kun Wang, Tianlong Chen, and Dawei Cheng. 2024b. G-designer: Architecting multi-agent communication topologies via graph neural networks. arXiv preprint arXiv:2410.11782. — [arXiv:2410.11782](https://arxiv.org/abs/2410.11782) find
  59. Jiaru Zou, Xiyuan Yang, Ruizhong Qiu, Gaotang Li, Katherine Tieu, Pan Lu, Ke Shen, Hanghang Tong, Yejin Choi, Jingrui He, James Zou, Mengdi Wang, and Ling Yang. 2025. Latent collaboration in multi-agent systems. arXiv preprint arXiv:2511.20639. — [arXiv:2511.20639](https://arxiv.org/abs/2511.20639) find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/sun-2026-topodim-one-shot-topology-generation/