Elamur Research⌕
☰
home / papers / 2025 / Can a Single Model Master Both Multi-turn Conversations and Tool Use? CoALM: A Unified Conversational Agentic Language Model

Can a Single Model Master Both Multi-turn Conversations and Tool Use? CoALM: A Unified Conversational Agentic Language Model

Paper · 2025 · Emre Can Acikgoz, Jeremiah Greer, Akul Datta et al.

Summary

Task-oriented dialogue (TOD) systems handle multi-turn conversations and track user goals well, but they are usually tied to a small fixed set of APIs. Language agents (LA) call tools well but tend to lose track of user intent across turns. The authors test models on MultiWOZ 2.4, BFCL V3 and API-Bank and confirm that specialized systems do well on one side and poorly on the other.

To close this gap, they build CoALM-IT, a multi-task instruction-tuning set with three parts:

  • dialogue state tracking data from SNIPS;
  • function-calling data from Hammer and ToolACE;
  • a new Conversational ReAct API (CRA) subset, derived from SGD and generated with GPT-4o, in which multi-turn dialogues contain explicit reasoning steps before API calls and before replies.

Fine-tuning Llama models of three sizes yields CoALM 8B, 70B and 405B. In zero-shot evaluation, the larger variants beat GPT-4o and domain-specific baselines on all three benchmarks. This shows that one open model can combine dialogue management and tool use. All code, weights, data and checkpoints are released.

Key points

  • Benchmarks: MultiWOZ 2.4 (TOD), BFCL V3 and API-Bank (LA). None of their training splits were used, so evaluation is zero-shot.
  • CoALM 405B reaches 100% on BFCL relevance detection and is reported as the strongest fully open-source model on the BFCL V3 leaderboard. Running it requires 16 H100 GPUs.
  • Ablations show each data component matters:
    • without LA data, API-Bank Rouge-L1 falls 47.3% and BFCL success falls 18.3%;
    • without DST data, JGA drops 11.0%;
    • without CRA data, MultiWOZ Success drops 11.7%.
  • No catastrophic forgetting was found for CoALM 8B on MMLU, ARC, GSM8K and HellaSwag; the average score moved from 70.87% to 70.93%.
  • Limitation: all experiments use only the Llama family.

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 Illinois Urbana-ChampaignUrbana · United States184 works · 69,372 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

🧭 Topics 2Research subjects and languages the work is about.Language models & pretraining · 3,320 worksDialogue systems · 624 works

Cite

Cite as Emre Can Acikgoz, Jeremiah Greer, Akul Datta, Ze Yang, William Zeng, Oussama Elachqar, Emmanouil Koukoumidis, Dilek Hakkani-Tür and Gokhan Tur. 2025. Can a Single Model Master Both Multi-turn Conversations and Tool Use? CoALM: A Unified Conversational Agentic Language Model. Proceedings of the 63rd 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/v1/2025.acl-long.605

@inproceedings{acikgoz2025can,
  title = {{Can a Single Model Master Both Multi-turn Conversations and Tool Use? CoALM: A Unified Conversational Agentic Language Model}},
  author = {Emre Can Acikgoz, Jeremiah Greer, Akul Datta, Ze Yang, William Zeng, Oussama Elachqar, Emmanouil Koukoumidis, Dilek Hakkani-Tür and Gokhan Tur},
  booktitle = {Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2025},
  doi = {10.18653/v1/2025.acl-long.605},
  url = {https://doi.org/10.18653/v1/2025.acl-long.605}
}

References

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

  1. Ibrahim Abdelaziz, Kinjal Basu, Mayank Agarwal, Sadhana Kumaravel, Matthew Stallone, Rameswar Panda, Yara Rizk, G P Shrivatsa Bhargav, Maxwell Crouse, Chulaka Gunasekara, Shajith Ikbal, Sachindra Joshi, Hima Karanam, Vineet Kumar, Asim Munawar, Sumit Neelam, Dinesh Raghu, Udit Sharma, Adriana Meza S… find
  2. BitsAndBytes. 2025. bitsandbytes. find
  3. Willy Chung, Samuel Cahyawijaya, Bryan Wilie, Holy Lovenia, and Pascale Fung. 2023. InstructTODS: Large language models for end-to-end task-oriented dialogue systems. In Proceedings of the Second Workshop on Natural Language Interfaces, pages 1– 21, Bali, Indonesia. Association for Computational Lin… find
  4. Peter Clark, Isaac Cowhey, Oren Etzioni, Tushar Khot, Ashish Sabharwal, Carissa Schoenick, and Oyvind Tafjord. 2018. Think you have solved question answering? try arc, the ai2 reasoning challenge. arXiv preprint arXiv:1803.05457.
  5. Karl Cobbe, Vineet Kosaraju, Mohammad Bavarian, Mark Chen, Heewoo Jun, Lukasz Kaiser, Matthias Plappert, Jerry Tworek, Jacob Hilton, Reiichiro Nakano, et al. 2021. Training verifiers to solve math word problems. arXiv preprint arXiv:2110.14168.
  6. Alice Coucke, Alaa Saade, Adrien Ball, Théodore Bluche, Alexandre Caulier, David Leroy, Clément Doumouro, Thibault Gisselbrecht, Francesco Caltagirone, Thibaut Lavril, et al. 2018. Snips voice platform: an embedded spoken language understanding system for private-by-design voice interfaces. arXiv pr…
  7. Tim Dettmers, Artidoro Pagnoni, Ari Holtzman, and Luke Zettlemoyer. 2023. Qlora: Efficient finetuning of quantized llms. Preprint, arXiv:2305.14314.
  8. Abhimanyu Dubey et al. 2024. The llama 3 herd of models. ArXiv, abs/2407.21783.
  9. Yujie Feng, Zexin Lu, Bo Liu, Liming Zhan, and Xiao-Ming Wu. 2023. Towards LLM-driven dialogue state tracking. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 739–755, Singapore. Association for Computational Linguistics. find
  10. Leo Gao, Jonathan Tow, Baber Abbasi, Stella Biderman, Sid Black, Anthony DiPofi, Charles Foster, Laurence Golding, Jeffrey Hsu, Alain Le Noac’h, Haonan Li, Kyle McDonell, Niklas Muennighoff, Chris Ociepa, Jason Phang, Laria Reynolds, Hailey Schoelkopf, Aviya Skowron, Lintang Sutawika, Eric Tang, Ani… find
  11. Prakhar Gupta, Cathy Jiao, Yi-Ting Yeh, Shikib Mehri, Maxine Eskenazi, and Jeffrey Bigham. 2022. In-structDial: Improving zero and few-shot generalization in dialogue through instruction tuning. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 505–525,… find
  12. Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, and Jacob Steinhardt. 2021. Measuring massive multitask language understanding. Proceedings of the International Conference on Learning Representations (ICLR).
  13. Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. 2021. Lora: Low-rank adaptation of large language models. arXiv preprint arXiv:2106.09685.
  14. Yushi Hu, Chia-Hsuan Lee, Tianbao Xie, Tao Yu, Noah A. Smith, and Mari Ostendorf. 2022. In-context learning for few-shot dialogue state tracking. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 2627–2643, Abu Dhabi, United Arab Emirates. Association for Computational…
  15. Vojtěch Hudeček and Ondrej Dusek. 2023. Are large language models all you need for task-oriented dialogue? In Proceedings of the 24th Annual Meeting of the Special Interest Group on Discourse and Dialogue, pages 216–228, Prague, Czechia. Association for Computational Linguistics. find
  16. Albert Q Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lucile Saulnier, et al. 2023. Mistral 7b. arXiv preprint arXiv:2310.06825. find
  17. Brendan King and Jeffrey Flanigan. 2024. Unsupervised end-to-end task-oriented dialogue with LLMs: The power of the noisy channel. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 8283–8300, Miami, Florida, USA. Association for Computational Linguistic… find
  18. Minghao Li, Yingxiu Zhao, Bowen Yu, Feifan Song, Hangyu Li, Haiyang Yu, Zhoujun Li, Fei Huang, and Yongbin Li. 2023. API-bank: A comprehensive benchmark for tool-augmented LLMs. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 3102–3116, Singapore. Ass…
  19. Zekun Li, Zhiyu Zoey Chen, Mike Ross, Patrick Huber, Seungwhan Moon, Zhaojiang Lin, Xin Luna Dong, Adithya Sagar, Xifeng Yan, and Paul A. Crook. 2024. Large language models as zero-shot dialogue state tracker through function calling. Preprint, arXiv:2402.10466. find
  20. Qiqiang Lin, Muning Wen, Qiuying Peng, Guanyu Nie, Junwei Liao, Jun Wang, Xiaoyun Mo, Jiamu Zhou, Cheng Cheng, Yin Zhao, and Weinan Zhang. 2024. Hammer: Robust function-calling for on-device language models via function masking. ArXiv, abs/2410.04587. — [arXiv:2410.04587](https://arxiv.org/abs/2410.04587) find
  21. Yuan Ling, Fanyou Wu, Shujing Dong, Yarong Feng, George Karypis, and Chandan K. Reddy. 2023. find
  22. International workshop on multimodal learning -2023 theme: Multimodal learning with foundation models. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD ’23, page 5868–5869, New York, NY, USA. Association for Computing Machinery. find
  23. Weiwen Liu, Xu Huang, Xingshan Zeng, Xinlong Hao, Shuai Yu, Dexun Li, Shuai Wang, Weinan Gan, Zhengying Liu, Yuanqing Yu, Zezhong Wang, Yuxian Wang, Wu Ning, Yutai Hou, Bin Wang, Chuhan Wu, Xinzhi Wang, Yong Liu, Yasheng Wang, Duyu Tang, Dandan Tu, Lifeng Shang, Xin Jiang, Ruiming Tang, Defu Lian, Q… find
  24. Marvin Minsky. 1986. The Society of Mind. Simon & Schuster. find
  25. OpenAI, Josh Achiam, et al. 2024. Gpt-4 technical report. Preprint, arXiv:2303.08774.
  26. Oumi. 2025. Oumi: an open, end-to-end platform for building large foundation models. find
  27. Bhargavi Paranjape, Scott Lundberg, Sameer Singh, Hannaneh Hajishirzi, Luke Zettlemoyer, and Marco Tulio Ribeiro. 2023. Art: Automatic multi-step reasoning and tool-use for large language models. arXiv preprint arXiv:2303.09014. find
  28. Shishir G Patil, Tianjun Zhang, Xin Wang, and Joseph E Gonzalez. 2023. Gorilla: Large language model connected with massive apis. arXiv preprint arXiv:2305.15334.
  29. Abhinav Rastogi, Xiaoxue Zang, Srinivas Sunkara, Raghav Gupta, and Pranav Khaitan. 2020. Towards scalable multi-domain conversational agents: The schema-guided dialogue dataset. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 34, pages 8689–8696.
  30. Timo Schick, Jane Dwivedi-Yu, Roberto Dessì, Roberta Raileanu, Maria Lomeli, Eric Hambro, Luke Zettlemoyer, Nicola Cancedda, and Thomas Scialom. 2024. Toolformer: Language models can teach themselves to use tools. Advances in Neural Information Processing Systems, 36.
  31. Yixuan Su, Lei Shu, Elman Mansimov, Arshit Gupta, Deng Cai, Yi-An Lai, and Yi Zhang. 2022. Multi-task pre-training for plug-and-play task-oriented dialogue system. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 4661–4676, Dub… find
  32. TogetherAI. 2025. Together ai: The ai acceleration cloud. find
  33. Marilyn A. Walker, Diane J. Litman, Candace A. Kamm, and Alicia Abella. 1997. PARADISE: A framework for evaluating spoken dialogue agents. In 35th Annual Meeting of the Association for Computational Linguistics and 8th Conference of the European Chapter of the Association for Computational Linguisti…
  34. Xingyao Wang, Yangyi Chen, Lifan Yuan, Yizhe Zhang, Yunzhu Li, Hao Peng, and Heng Ji. 2024. Executable code actions elicit better llm agents. In ICLR 2024 Workshop on Large Language Model (LLM) Agents.
  35. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, brian ichter, Fei Xia, Ed Chi, Quoc V Le, and Denny Zhou. 2022. Chain-of-thought prompting elicits reasoning in large language models. In Advances in Neural Information Processing Systems, volume 35, pages 24824–24837. Curran Associates, Inc.
  36. Heng-Da Xu, Xian-Ling Mao, Puhai Yang, Fanshu Sun, and Heyan Huang. 2024. Rethinking task-oriented dialogue systems: From complex modularity to zero-shot autonomous agent. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2748–2… find
  37. Fanjia Yan, Huanzhi Mao, Charlie Cheng-Jie Ji, Tianjun Zhang, Shishir G. Patil, Ion Stoica, and Joseph E. Gonzalez. 2024. Berkeley function calling leaderboard. https://gorilla.cs.berkeley.edu/blogs/8_berkeley_function_calling_ leaderboard.html. ↗
  38. Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, and Yuan Cao. 2023. React: Synergizing reasoning and acting in language models. Preprint, arXiv:2210.03629.
  39. Fanghua Ye, Jarana Manotumruksa, and Emine Yilmaz. 2022. MultiWOZ 2.4: A multi-domain task-oriented dialogue dataset with essential annotation corrections to improve state tracking evaluation. In Proceedings of the 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue, pages 35…
  40. Steve Young. 2002. Talking to machines (statistically speaking). In 7th International Conference on Spoken Language Processing (ICSLP 2002), pages 9–16. find
  41. Rowan Zellers, Ari Holtzman, Yonatan Bisk, Ali Farhadi, and Yejin Choi. 2019. HellaSwag: Can a machine really finish your sentence? In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 4791–4800, Florence, Italy. Association for Computational Linguistics. find
  42. Jianguo Zhang, Tian Lan, Ming Zhu, Zuxin Liu, Thai Hoang, Shirley Kokane, Weiran Yao, Juntao Tan, Akshara Prabhakar, Haolin Chen, et al. 2024. xlam: A family of large action models to empower ai agent systems. CoRR.
  43. Xiaoying Zhang, Baolin Peng, Kun Li, Jingyan Zhou, and Helen Meng. 2023. SGP-TOD: Building task bots effortlessly via schema-guided LLM prompting. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 13348–13369, Singapore. Association for Computational Linguistics. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/acikgoz-2025-can-single-model-master-both/