Elamur Research⌕
☰
home / papers / 2026 / MUDiC: A Dataset for Multi-User Dialogue and Collaboration in Chatbot Interaction

MUDiC: A Dataset for Multi-User Dialogue and Collaboration in Chatbot Interaction

Summary

Most dialogue corpora record one person talking to a chatbot. Many real tasks, however, involve several people coordinating through a single system, and the shortage of authentic group data has slowed progress on such agents. The authors address this gap with MUDiC, a corpus of real conversations rather than simulated or rewritten ones. In each session, a pair of users and a chatbot negotiate the place, date and time of a meeting in a Telegram group chat. The bot, built with Rasa, tracked each user's state and the group's state. It followed one of four dialogue policies, which differ in when the bot takes the turn and how it offers help. These range from staying silent until someone asks for help to actively reporting availability information. The released data contain 1,689 exchanges from 20 groups, annotated with user IDs, turns, intents, dialogue acts, extracted slot values and logs of the system's decisions. RoBERTa-based models were also used to label utterance language and to analyse speaker transitions. The resource supports work on modelling conversations with several users, resolving ambiguous intents and studying how a bot moderates group interaction.

Key points

  • MUDiC contains 1,689 dialogue exchanges between 20 two-person groups and a task-oriented chatbot. It is publicly available on Zenodo (doi:10.5281/zenodo.19037937).
  • The infrastructure combined Telegram, accessed via its API, with the Rasa framework for NLU and dialogue management. Policies relied on the Transformer Embedding Dialogue (TED) model, with a decision tree as a fallback for exceptional cases, and the system was hosted on a cloud VM with Apache2.
  • Four dialogue policies controlled the bot's turn-taking and assistance style. Examples include a passive Baseline that responds only to explicit "Help" requests and a Notification policy that reacts to mentions of location, date or time.
  • The intent inventory covers several categories:
    • session control (start, stop, greet, goodbye);
    • agreement (affirm, deny);
    • scheduling (e.g., date_suggestion, earlier, later).
  • Stated limitations:
    • the dialogues are in English, but participants were mostly native German speakers;
    • the task scenario was fixed;
    • the dataset is small because group sessions were hard to organise.
  • Planned future work includes groups of three or more users, additional languages and domains, and combining policy-based control with LLMs.

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

Fields: Linguistics & languages

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.63317/36zqrpocuvn9

@inproceedings{wagner2026mudic,
  title = {{MUDiC: A Dataset for Multi-User Dialogue and Collaboration in Chatbot Interaction}},
  author = {Wagner, Nicolas and Jimenez, Cristina Luna and Andre, Elisabeth and Minker, Wolfgang and Ultes, Stefan},
  booktitle = {Proceedings of the Fifteenth Language Resources and Evaluation Conference},
  year = {2026},
  doi = {10.63317/36zqrpocuvn9},
  url = {https://doi.org/10.63317/36zqrpocuvn9}
}

References

19 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Eleni Adamopoulou and Lefteris Moussiades. 2020. An overview of chatbot technology. In IFIP international conference on artificial intelligence applications and innovations, pages 373–383. Springer. find
  2. Paweł Budzianowski, Tsung-Hsien Wen, Bo-Hsiang Tseng, Iñigo Casanueva, Stefan Ultes, Osman Ramadan, and Milica Gašić. 2018. Multi-WOZ - a large-scale multi-domain Wizard-of-Oz dataset for task-oriented dialogue modelling. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language… find
  3. Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer, and Veselin Stoyanov. 2020. Unsupervised cross-lingual representation learning at scale. In Proceedings of the 58th Annual Meeting of the Association fo… find
  4. Ilaria Fiorentini, Marco Forlano, and Nicholas Nese. 2024. Towards the WhAP corpus: A resource for the study of Italian on WhatsApp. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 16659–16663, Torino… find
  5. Ananya Ganesh, Martha Palmer, and Katharina Kann. 2023. A survey of challenges and methods in the computational modeling of multi-party dialog. In Proceedings of the 5th Workshop on NLP for Conversational AI (NLP4ConvAI 2023), pages 140–154, Toronto, Canada. Association for Computational Linguistics… find
  6. Yohan Jo, Xinyan Zhao, Arijit Biswas, Nikoletta Basiou, Vincent Auvray, Nikolaos Malandrakis, Angeliki Metallinou, and Alexandros Potamianos. 2023. Multi-user MultiWOZ: Task-oriented dialogues among multiple users. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 3237–… find
  7. Matthias Kraus, Stina Klein, Nicolas Wagner, Wolfgang Minker, and Elisabeth André. 2024. A pilot study on multi-party conversation strategies for group recommendations. In Proceedings of the 6th ACM Conference on Conversational User Interfaces, CUI ’24, New York, NY, USA. Association for Computing M… find
  8. Mohammad Amin Kuhail, Imran Taj, Saifeddin Al-imamy, and Bayan Abu Shawar. 2024. A review on polyadic chatbots: trends, challenges, and future research directions. Knowl. Inf. Syst., 67(1):109–165. find
  9. Han Li, Renwen Zhang, Yi-Chieh Lee, Robert E Kraut, and David C Mohr. 2023. Systematic review and meta-analysis of ai-based conversational agents for promoting mental health and well-being. NPJ Digital Medicine, 6(1):236. find
  10. Cristina Luna-Jiménez, Manuel Gil-Martín, Luis Fernando D’Haro, Fernando Fernández-Martínez, and Rubén San-Segundo. 2024. Evaluating emotional and subjective responses in synthetic art-related dialogues: A multi-stage framework with large language models. Expert Systems with Applications, 255:124524… find
  11. Khyati Mahajan and Samira Shaikh. 2021. On the need for thoughtful data collection for multi-party dialogue: A survey of available corpora and collection methods. In Proceedings of the 22nd annual meeting of the special interest group on discourse and dialogue, pages 338–352. find
  12. Danila Mamontov, Alexey Karpov, and Wolfgang Minker. 2025. A multilingual telegram chatbot for mental health data collection. In Proceedings of the 27th International Conference on Multimodal Interaction, pages 794–796. find
  13. Manqing Mao, Paishun Ting, Yijian Xiang, Mingyang Xu, Julia Chen, and Jianzhe Lin. 2024. Multi-user chat assistant (muca): a framework using llms to facilitate group conversations. find
  14. Michael McTear. 2022. Conversational ai: Dialogue systems, conversational agents, and chatbots. Springer Nature. find
  15. María Jesús Rodríguez-Sánchez, Zoraida Callejas, Angel Ruiz-Zafra, and Kawtar Benghazi. 2025. Breaking the limits of chatbot development: Api-driven multi-domain chatbot generation empowered by generative ai. Computing, 107(11):1–32. find
  16. Sonali Uttam Singh and Akbar Siami Namin. 2025. A survey on chatbots and large language models: Testing and evaluation techniques. Natural Language Processing Journal, 10:100128. find
  17. Vladimir Vlasov, Johannes E. M. Mosig, and Alan Nichol. 2019. Dialogue transformers. CoRR, abs/1910.00486. — [arXiv:1910.00486](https://arxiv.org/abs/1910.00486) find
  18. Nicolas Wagner, Matthias Kraus, Wolfgang Minker, David Griol, and Zoraida Callejas. 2025. A survey on multi-user conversational interfaces. Applied Sciences, 15(13):7267. find
  19. Nicolas Wagner, Matthias Kraus, Tibor Tonn, and Wolfgang Minker. 2022. Comparing moderation strategies in group chats with multi-user chatbots. In Proceedings of the 4th Conference on Conversational User Interfaces, CUI ’22, New York, NY, USA. Association for Computing Machinery. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/wagner-2026-mudic-dataset-multi-user-dialogue/