Elamur Research⌕
☰
home / papers / 2026 / TeamFusion: Supporting Open-ended Teamwork with Multi-Agent Systems

TeamFusion: Supporting Open-ended Teamwork with Multi-Agent Systems

Paper · 2026 · Jiale Liu, Victor Bursztyn, Lin Ai et al.

Summary

Many team decisions have no single right answer, and the usual way of using LLMs here, merging everyone's input in one pass, tends to blur conflicts and silence minority opinions. TeamFusion is a multi-agent framework built to avoid this. It works in four stages. First, each team member gets a proxy agent, personalized in-context from that person's stated preferences. Second, the proxies hold a round-robin group-chat discussion that brings agreements and conflicts into the open. Third, a remix agent turns the transcript and task context into an editable deliverable. Fourth, the deliverable can go back into discussion for further refinement. The authors also propose an asynchronous human-in-the-loop evaluation protocol that gathers preferences separately from the interaction, so studies with professional experts can scale. Across two teamwork tasks (text and multimodal), several backbone models, and different team sizes, TeamFusion beat direct-aggregation baselines. It also increased agreement among participants. The work suggests that simulating deliberation, rather than averaging inputs, produces more representative and auditable group outputs.

Key points

  • The pipeline has four phases: Represent, Discuss, Remix, and Critique/Refine. Discussion uses a deterministic round-robin protocol inspired by divergence–convergence models and the nominal group technique.
  • After exposure to TeamFusion outputs, team agreement measured by mean Kendall's W rose from 0.37 to 0.43, moving from "Fair" to "Moderate" agreement.
  • Designers rated how well proxy comments represented them at a mean of 4.06 (σ = 1.07). Over 75% of comments were rated positively, and the lowest per-designer mean was 3.53 (σ across designers = 0.29).
  • A live pilot used a counterbalanced crossover design with six participants in two teams of three. Teams using TeamFusion reached decisions faster than in free-form discussion and reported better representativeness, clearer trade-offs, and higher satisfaction. Most participants preferred TeamFusion.
  • The main limitation is that the system assumes a flat team hierarchy and ignores roles such as art directors or clients, as well as seniority.

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

🎓Columbia UniversityNew York · United States241 works · 54,053 citations🎓Pennsylvania State UniversityState College · United States56 works · 5,360 citations🏢Adobe SystemsSan Jose · United States35 works · 3,711 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 works

Cite

Cite as Jiale Liu, Victor Bursztyn, Lin Ai, Haoliang Wang, Sunav Choudhary, Saayan Mitra and Qingyun Wu. 2026. TeamFusion: Supporting Open-ended Teamwork with Multi-Agent Systems. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.acl-long.657

@inproceedings{liu2026teamfusion,
  title = {{TeamFusion: Supporting Open-ended Teamwork with Multi-Agent Systems}},
  author = {Jiale Liu, Victor Bursztyn, Lin Ai, Haoliang Wang, Sunav Choudhary, Saayan Mitra and Qingyun Wu},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.657},
  url = {https://doi.org/10.18653/v1/2026.acl-long.657}
}

References

61 references, 13 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Selcuk Acar and Mark A Runco. 2019. Divergent thinking: New methods, recent research, and extended theory. Psychology of aesthetics, creativity, and the arts, 13(2):153. find
  2. Divyansh Agarwal, Alexander Fabbri, Ben Risher, Philippe Laban, Shafiq Joty, and Chien-Sheng Wu. 2024. Prompt leakage effect and mitigation strategies for multi-turn LLM applications. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing: Industry Track, pages 125… find
  3. Adithya Bhaskar, Alex Fabbri, and Greg Durrett. 2023. Prompted opinion summarization with GPT-3.5. In John Chen, Xi Lu, Yuzhou Du, Michael Rejtig, Ruth Bagley, Mike Horn, and Uri Wilensky. 2024. Learning agent-based modeling with llm companions: Experiences of novices and experts using chatgpt & net… find
  4. Steffi Chern, Zhen Fan, and Andy Liu. 2024. Combating adversarial attacks with multi-agent debate. arXiv preprint arXiv:2401.05998. — [arXiv:2401.05998](https://arxiv.org/abs/2401.05998) find
  5. Chun-Wei Chiang, Zhuoran Lu, Zhuoyan Li, and Ming Yin. 2024. Enhancing ai-assisted group decision making through llm-powered devil’s advocate. In Proceedings of the 29th International Conference on Intelligent User Interfaces, pages 103–119. find
  6. Luis Fariñas del Cerro, Andreas Herzig, Dominique Longin, and Omar Rifi. 1998. Belief reconstruction in cooperative dialogues. In International Conference on Artificial Intelligence: Methodology, Systems, and Applications, pages 254–266. Springer. find
  7. Karen L. Dowling and Robert D. St. Louis. 2000. Asynchronous implementation of the nominal group technique: is it effective? Decis. Support Syst., 29(3):229–248. find
  8. 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.
  9. B Aubrey Fisher. 1970. Decision emergence: Phases in group decision-making. Communications Monographs, 37(1):53–66. find
  10. Jarod Govers, Eduardo Velloso, Vassilis Kostakos, and Jorge Goncalves. 2024. Ai-driven mediation strategies for audience depolarisation in online debates. In Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems, CHI ’24, New York, NY, USA. Association for Computing Machinery. find
  11. Kai Greshake, Sahar Abdelnabi, Shailesh Mishra, Christoph Endres, Thorsten Holz, and Mario Fritz. 2023. Not what you’ve signed up for: Compromising real-world llm-integrated applications with indirect prompt injection. In Proceedings of the 16th ACM workshop on artificial intelligence and security,…
  12. Xudong Guo, Kaixuan Huang, Jiale Liu, Wenhui Fan, Natalia Vélez, Qingyun Wu, Huazheng Wang, Thomas L Griffiths, and Mengdi Wang. 2024. Embodied llm agents learn to cooperate in organized teams. arXiv preprint arXiv:2403.12482. — [arXiv:2403.12482](https://arxiv.org/abs/2403.12482) find
  13. Shanshan Han, Qifan Zhang, Weizhao Jin, and Zhaozhuo Xu. 2024. Llm multi-agent systems: Challenges and open problems. arXiv preprint arXiv:2402.03578. — [arXiv:2402.03578](https://arxiv.org/abs/2402.03578) find
  14. Sara Kiesler and Lee Sproull. 1992. Group decision making and communication technology. Organizational behavior and human decision processes, 52(1):96–123. find
  15. Yubin Kim, Chanwoo Park, Hyewon Jeong, Yik S Chan, Xuhai Xu, Daniel McDuff, Hyeonhoon Lee, Marzyeh Ghassemi, Cynthia Breazeal, and Hae W Park. 2024. Mdagents: An adaptive collaboration of llms for medical decision-making. Advances in Neural Information Processing Systems, 37:79410–79452. find
  16. Alfred Kobsa. 1989. A taxonomy of beliefs and goals for user models in dialog systems. In User models in dialog systems, pages 52–68. Springer. find
  17. Kenneth L Kraemer and John Leslie King. 1988. Computer-based systems for cooperative work and group decision making. ACM Computing Surveys (CSUR), 20(2):115–146. find
  18. D. Kwon, Sunwoo Lee, Ki Hyun Kim, Seojin Lee, Tae-Yoon Kim, and Eric Davis. 2023. What, when, and how to ground: Designing user persona-aware conversational agents for engaging dialogue. In Annual Meeting of the Association for Computational Linguistics. find
  19. 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.
  20. Philippe Laban, Wojciech Kryściński, Divyansh Agarwal, Alexander Richard Fabbri, Caiming Xiong, Shafiq Joty, and Chien-Sheng Wu. 2023. Summedits: Measuring llm ability at factual reasoning through the lens of summarization. In Proceedings of the 2023 conference on empirical methods in natural lang… find
  21. Zhe Hu, Hou Pong Chan, Jing Li, and Yu Yin. 2025. Debate-to-write: A persona-driven multi-agent framework for diverse argument generation. In Proceedings of the 31st International Conference on Computational Linguistics, pages 4689–4703.
  22. Dawei Li, Bohan Jiang, Liangjie Huang, Alimohammad Beigi, Chengshuai Zhao, Zhen Tan, Amrita Bhattacharjee, Yuxuan Jiang, Canyu Chen, Tianhao Wu, and 1 others. 2025. From generation to judgment: Opportunities and challenges of llm-as-a-judge. In Proceedings of the 2025 Conference on Empirical Methods… find
  23. Lei Huang, Weijiang Yu, Weitao Ma, Weihong Zhong, Zhangyin Feng, Haotian Wang, Qianglong Chen, Weihua Peng, Xiaocheng Feng, Bing Qin, and 1 others. 2025. A survey on hallucination in large language models: Principles, taxonomy, challenges, and open questions. ACM Transactions on Information Systems,…
  24. Nannan Huang, Lin Tian, Haytham Fayek, and Xiuzhen Zhang. 2023. Examining bias in opinion summarisation through the perspective of opinion diversity. In Proceedings of the 13th Workshop on Computational Approaches to Subjectivity, Sentiment, & Social Media Analysis, pages 149–161, Toronto, Canada. A… find
  25. Qiushi Huang, Xubo Liu, Tom Ko, Boyong Wu, Wenwu Wang, Yu Zhang, and Lilian Tang. 2024. Selective prompting tuning for personalized conversations with llms. ArXiv, abs/2406.18187. find
  26. Pontus Johansson. 2002. User modeling in dialog systems. St. Anna Report SAR, pages 02–2. find
  27. Miao Li, Eduard Hovy, and Jey Lau. 2023. Summarizing multiple documents with conversational structure for meta-review generation. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 7089–7112, Singapore. Association for Computational Linguistics. find
  28. Miao Li, Jey Han Lau, and Eduard Hovy. 2024. A sentiment consolidation framework for meta-review generation. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 10158–10177, Bangkok, Thailand. Association for Computational Linguis… find
  29. Tian Liang, Zhiwei He, Wenxiang Jiao, Xing Wang, Yan Wang, Rui Wang, Yujiu Yang, Shuming Shi, and Zhaopeng Tu. 2024. Encouraging divergent thinking in large language models through multi-agent debate. In Proceedings of the 2024 conference on empirical methods in natural language processing, pages 17…
  30. Xiaogeng Liu, Zhiyuan Yu, Yizhe Zhang, Ning Zhang, and Chaowei Xiao. 2024. Automatic and universal prompt injection attacks against large language models. arXiv preprint arXiv:2403.04957. — [arXiv:2403.04957](https://arxiv.org/abs/2403.04957) find
  31. Yang Liu, Yuanshun Yao, Jean-Francois Ton, Xiaoying Zhang, Ruocheng Guo, Hao Cheng, Yegor Klochkov, Muhammad Faaiz Taufiq, and Hang Li. 2023. Trustworthy llms: a survey and guideline for evaluating large language models’ alignment. arXiv preprint arXiv:2308.05374.
  32. Aman Madaan, Niket Tandon, Prakhar Gupta, Skyler Hallinan, Luyu Gao, Sarah Wiegreffe, Uri Alon, Nouha Dziri, Shrimai Prabhumoye, Yiming Yang, and 1 others. 2023. Self-refine: Iterative refinement with self-feedback. Advances in Neural Information Processing Systems, 36:46534–46594. find
  33. Humza Naveed, Asad Ullah Khan, Shi Qiu, Muhammad Saqib, Saeed Anwar, Muhammad Usman, Naveed Akhtar, Nick Barnes, and Ajmal Mian. 2025. A comprehensive overview of large language models. ACM Transactions on Intelligent Systems and Technology, 16(5):1–72.
  34. R Orwig, Hsinchun Chen, D Vogel, and Jay F Nunamaker. 1997. A multi-agent view of strategic planning using group support systems and artificial intelligence. Group Decision and Negotiation, 6(1):37–59. find
  35. Letitia Parcalabescu and Anette Frank. 2024. On measuring faithfulness or self-consistency of natural language explanations. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 6048– 6089. find
  36. Joon Sung Park, Joseph O’Brien, Carrie Jun Cai, Meredith Ringel Morris, Percy Liang, and Michael S Bernstein. 2023. Generative agents: Interactive simulacra of human behavior. In Proceedings of the 36th annual acm symposium on user interface software and technology, pages 1–22. find
  37. Giorgio Piatti, Zhijing Jin, Max Kleiman-Weiner, Bernhard Schölkopf, Mrinmaya Sachan, and Rada Mihalcea. 2024. Cooperate or collapse: Emergence of sustainable cooperation in a society of llm agents. Advances in Neural Information Processing Systems, 37:111715–111759. find
  38. Steven Rogelberg, Desmond J. Leach, Peter B. Warr, and Jennifer L. Burnfield. 2006. "not another meeting!" are meeting time demands related to employee well-being? The Journal of applied psychology, 91 1:83–96. find
  39. Alexander C. Romney, Joseph A. Allen, and Zahra Hey-darifard. 2025. Meeting load paradox: Balancing the benefits and burdens of work meetings. Business Horizons, 68(1):33–43. find
  40. Mark A Runco and Selcuk Acar. 2012. Divergent thinking as an indicator of creative potential. Creativity research journal, 24(1):66–75. find
  41. Donald G. Saari. 1995. Basic Geometry of Voting, 1 edition. Springer-Verlag Berlin Heidelberg. find
  42. Mrinank Sharma, Meg Tong, Tomasz Korbak, David Duvenaud, Amanda Askell, Samuel R Bowman, Esin DURMUS, Zac Hatfield-Dodds, Scott R Johnston, Shauna M Kravec, and 1 others. Towards understanding sycophancy in language models. In The Twelfth International Conference on Learning Representations.
  43. Lin Shi, Chiyu Ma, Wenhua Liang, Xingjian Diao, Weicheng Ma, and Soroush Vosoughi. 2024. Judging the judges: A systematic study of position bias in llm-as-a-judge. arXiv preprint arXiv:2406.07791.
  44. Joon Gi Shin, Janin Koch, Andrés Lucero, Peter Dalsgaard, and Wendy E. Mackay. 2023. Integrating ai in human-human collaborative ideation. In Extended Abstracts of the 2023 CHI Conference on Human Factors in Computing Systems, CHI EA ’23, New York, NY, USA. Association for Computing Machinery. find
  45. Elizabeth Shriberg, Raj Dhillon, Sonali Bhagat, Jeremy Ang, and Hannah Carvey. 2004. The ICSI meeting recorder dialog act (MRDA) corpus. In Proceedings of the 5th SIGdial Workshop on Discourse and Dialogue at HLT-NAACL 2004, pages 97–100, Cambridge, Massachusetts, USA. Association for Computational…
  46. Linxin Song, Jiale Liu, Jieyu Zhang, Shaokun Zhang, Ao Luo, Shijian Wang, Qingyun Wu, and Chi Wang. 2024. Adaptive in-conversation team building for language model agents. arXiv preprint arXiv:2405.19425. — [arXiv:2405.19425](https://arxiv.org/abs/2405.19425) find
  47. Qiushi Sun, Zhangyue Yin, Xiang Li, Zhiyong Wu, Xipeng Qiu, and Lingpeng Kong. Corex: Pushing the boundaries of complex reasoning through multi-model collaboration. In First Conference on Language Modeling. find
  48. Michael Henry Tessler, Michiel A. Bakker, Daniel Jarrett, Hannah Sheahan, Martin J. Chadwick, Raphael Koster, Georgina Evans, Lucy Campbell-Gillingham, Tantum Collins, David C. Parkes, Matthew Botvinick, and Christopher Summerfield. 2024. Ai can help humans find common ground in democratic deliberat… find
  49. Khanh-Tung Tran, Dung Dao, Minh-Duong Nguyen, Quoc-Viet Pham, Barry O’Sullivan, and Hoang D Nguyen. 2025. Multi-agent collaboration mechanisms: A survey of llms. arXiv preprint arXiv:2501.06322. find
  50. Cunxiang Wang, Xiaoze Liu, Yuanhao Yue, Xiangru Tang, Tianhang Zhang, Cheng Jiayang, Yunzhi Yao, Wenyang Gao, Xuming Hu, Zehan Qi, and 1 others. 2023. Survey on factuality in large language models: Knowledge, retrieval and domain-specificity. arXiv preprint arXiv:2310.07521.
  51. Zheng Wang, Zhongyang Li, Zeren Jiang, Dandan Tu, and Wei Shi. 2024. Crafting personalized agents through retrieval-augmented generation on editable memory graphs. In Conference on Empirical Methods in Natural Language Processing. find
  52. 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.
  53. Qingyun Wu, Gagan Bansal, Jieyu Zhang, Yiran Wu, Beibin Li, Erkang Zhu, Li Jiang, Xiaoyun Zhang, Shaokun Zhang, Jiale Liu, Ahmed Hassan Awadallah, Ryen W White, Doug Burger, and Chi Wang. 2024. Autogen: Enabling next-gen LLM applications via multi-agent conversations. In First Conference on Language… find
  54. Kota Yamaguchi. 2021. Canvasvae: Learning to generate vector graphic documents. ICCV. find
  55. Miao Yu, Fanci Meng, Xinyun Zhou, Shilong Wang, Junyuan Mao, Linsey Pan, Tianlong Chen, Kun Wang, Xinfeng Li, Yongfeng Zhang, and 1 others. 2025. A survey on trustworthy llm agents: Threats and countermeasures. In Proceedings of the 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining V… find
  56. Haopeng Zhang, Philip S Yu, and Jiawei Zhang. 2025. A systematic survey of text summarization: From statistical methods to large language models. ACM Computing Surveys, 57(11):1–41. find
  57. Shaokun Zhang, Jieyu Zhang, Jiale Liu, Linxin Song, Chi Wang, Ranjay Krishna, and Qingyun Wu. 2024a. Offline training of language model agents with functions as learnable weights. In Forty-first International Conference on Machine Learning. find
  58. Shaowei Zhang and Deyi Xiong. 2025. Debate4math: Multi-agent debate for fine-grained reasoning in math. In Findings of the Association for Computational Linguistics: ACL 2025, pages 16810–16824. find
  59. Yang Zhang, Hanlei Jin, Dan Meng, Jun Wang, and Jinghua Tan. 2024b. A comprehensive survey on process-oriented automatic text summarization with exploration of llm-based methods. arXiv preprint arXiv:2403.02901. — [arXiv:2403.02901](https://arxiv.org/abs/2403.02901) find
  60. Yusen Zhang, Nan Zhang, Yixin Liu, Alexander Fabbri, Junru Liu, Ryo Kamoi, Xiaoxin Lu, Caiming Xiong, Jieyu Zhao, Dragomir Radev, Kathleen McKeown, and Rui Zhang. 2024c. Fair abstractive summarization of diverse perspectives. In Proceedings of the 2024 Conference of the North American Chapter of the… find
  61. Shenzhe Zhu, Shu Yang, Michiel A Bakker, Alex Pentland, and Jiaxin Pei. 2025. Can ai truly represent your voice in deliberations? a comprehensive study of large-scale opinion aggregation with llms. arXiv preprint arXiv:2510.05154. — [arXiv:2510.05154](https://arxiv.org/abs/2510.05154) find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/liu-2026-teamfusion-supporting-open-ended-teamwork/