Elamur Research⌕
☰
home / papers / 2026 / RECAP: REwriting Conversations for Intent Understanding in Agentic Planning

RECAP: REwriting Conversations for Intent Understanding in Agentic Planning

Paper · 2026 · Kushan Mitra, Dan Zhang, Hannah Kim et al.

Summary

Conversational assistants that coordinate several LLM agents need a reliable picture of what the user actually wants. Real dialogues are messy: requests can be vague, goals change midway, off-topic chatter intrudes, and several aims get mixed together. Classifiers built on a fixed schema of intents handle this open-ended input poorly.

The authors propose intent rewriting instead. A rewriter condenses a USER-AGENT dialogue into a short, current statement of the user's goal, which a planner then turns into a DAG of sub-tasks. They release RECAP, a benchmark of synthetic dialogues generated with GPT-4o and LLaMA 3.3-70B and checked by humans. It covers five topics, three length levels and several difficulty types. They also build an LLM-based judge that compares plans by their usefulness.

A carefully prompted rewriter beats baseline approaches in plan preference. Two DPO-tuned rewriters improve on it further, and the best one is equal or better in 77.8% of comparisons. The work presents rewriting as a practical, planner-agnostic way to improve agentic planning.

Key points

  • RECAP contains 810 human-validated conversations. Each must end on a user turn, and dialogues where the agent hallucinated or tried to solve the task were removed.
  • The dataset is organised along three axes:
    • Topics: cooking, programming, health, flights, restaurants.
    • Lengths: short, medium and long, averaging about 3, 7 and 12 utterances.
    • Intent categories: shifted, noisy, underspecified, multi-intent and perfect intent.
  • Plans from raw dialogue history differ noticeably from plans built on rewrites. This holds across several planner LLMs, including o3-mini.
  • Two DPO-based rewriters were fine-tuned starting from the strongest zero-shot prompt rewriter. The best one reaches a combined win-plus-tie rate of 77.8%.
  • Code is released at github.com/megagonlabs/recap and data on Hugging Face as megagonlabs/recap.
  • The stated limitations are:
    • inputs are text-only;
    • the structural metrics may miss plans that differ in form but work the same way;
    • rewrites are not directly optimised for plan structure.

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, 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 Kushan Mitra, Dan Zhang, Hannah Kim and Estevam Hruschka. 2026. RECAP: REwriting Conversations for Intent Understanding in Agentic Planning. Findings of the Association for Computational Linguistics: EACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.findings-eacl.105

@inproceedings{mitra2026recap,
  title = {{RECAP: REwriting Conversations for Intent Understanding in Agentic Planning}},
  author = {Kushan Mitra, Dan Zhang, Hannah Kim and Estevam Hruschka},
  booktitle = {Findings of the Association for Computational Linguistics: EACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-eacl.105},
  url = {https://doi.org/10.18653/v1/2026.findings-eacl.105}
}

References

31 references, 8 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Gaurav Arora, Shreya Jain, and Srujana Merugu. 2024. Intent detection in the age of LLMs. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing: Industry Track, pages 1559–1570, Miami, Florida, US. Association for Computational Linguistics. find
  2. Paweł Budzianowski, Tsung-Hsien Wen, Bo-Hsiang Tseng, Iñigo Casanueva, Stefan Ultes, Osman Ramadan, and Milica Gašić. 2018. MultiWOZ - 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…
  3. Bill Byrne, Karthik Krishnamoorthi, Chinnadhurai Sankar, Arvind Neelakantan, Ben Goodrich, Daniel Duckworth, Semih Yavuz, Amit Dubey, Kyu-Young Kim, and Andy Cedilnik. 2019. Taskmaster-1: Toward a realistic and diverse dialog dataset. In Proceedings of the 2019 Conference on Empirical Methods in Nat…
  4. Maximillian Chen, Ruoxi Sun, Tomas Pfister, and Sercan Ö. Arık. 2025. Learning to clarify: Multi-turn conversations with action-based contrastive self-training. find
  5. Yang Deng, Xuan Zhang, Wenxuan Zhang, Yifei Yuan, See-Kiong Ng, and Tat-Seng Chua. 2024. On the multi-turn instruction following for conversational web agents. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 8795–8812, Bangkok… find
  6. Lutfi Eren Erdogan, Hiroki Furuta, Sehoon Kim, Nicholas Lee, Suhong Moon, Gopala Anumanchipalli, Kurt Keutzer, and Amir Gholami. 2025. Plan-and-act: Improving planning of agents for long-horizon tasks. In Forty-second International Conference on Machine Learning. find
  7. Elnara Galimzhanova, Cristina Ioana Muntean, Franco Maria Nardini, Raffaele Perego, and Guido Rocchietti. 2023. Rewriting conversational utterances with instructed large language models. 2023 IEEE International Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT), pages 56–63. find
  8. Bingzheng Gan, Yufan Zhao, Tianyi Zhang, Jing Huang, Li Yusu, Shu Xian Teo, Changwang Zhang, and Wei Shi. 2025. MASTER: A multi-agent system with LLM specialized MCTS. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human… find
  9. Chih-Wen Goo, Guang Gao, Yun-Kai Hsu, Chih-Li Huo, Tsung-Chieh Chen, Keng-Wei Hsu, and Yun-Nung Chen. 2018. Slot-gated modeling for joint slot filling and intent prediction. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human La…
  10. Seungone Kim, Juyoung Suk, Xiang Yue, Vijay Viswanathan, Seongyun Lee, Yizhong Wang, Kiril Gashteovski, Carolin Lawrence, Sean Welleck, and Graham Neubig. 2025. Evaluating language models as synthetic data generators. In Proceedings of the 63rd Annual Meeting of the Association for Computational Lin…
  11. 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
  12. Ao Li, Yuexiang Xie, Songze Li, Fugee Tsung, Bolin Ding, and Yaliang Li. 2025. Agent-oriented planning in multi-agent systems. In The Thirteenth International Conference on Learning Representations. find
  13. Jieyu Li, Zhi Chen, Lu Chen, Zichen Zhu, Hanqi Li, Ruisheng Cao, and Kai Yu. 2023. Dir: A large-scale dialogue rewrite dataset for cross-domain conversational text-to-sql. Applied Sciences. find
  14. Gaurav Maheshwari, Dmitry Ivanov, and Kevin El Haddad. 2024. Efficacy of synthetic data as a benchmark. find
  15. Fengran Mo, Abbas Ghaddar, Kelong Mao, Mehdi Rezagholizadeh, Boxing Chen, Qun Liu, and Jian-Yun Nie. 2024. CHIQ: Contextual history enhancement for improving query rewriting in conversational search. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 225… find
  16. Nikola Mrkšić, Diarmuid Ó Séaghdha, Tsung-Hsien Wen, Blaise Thomson, and Steve Young. 2017. Neural belief tracker: Data-driven dialogue state tracking. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1777–1788, Vancouver, Can… find
  17. Cheng Qian, Bingxiang He, Zhong Zhuang, Jia Deng, Yujia Qin, Xin Cong, Zhong Zhang, Jie Zhou, Yankai Lin, Zhiyuan Liu, and Maosong Sun. 2024. Tell me more! towards implicit user intention understanding of language model driven agents. In Proceedings of the 62nd Annual Meeting of the Association for… find
  18. Shuofei Qiao, Runnan Fang, Zhisong Qiu, Xiaobin Wang, Ningyu Zhang, Yong Jiang, Pengjun Xie, Fei Huang, and Huajun Chen. 2025. Benchmarking agentic workflow generation. find
  19. Rafael Rafailov, Archit Sharma, Eric Mitchell, Stefano Ermon, Christopher D. Manning, and Chelsea Finn. 2023. Direct preference optimization: your language model is secretly a reward model. In Proceedings of the 37th International Conference on Neural Information Processing Systems, NIPS ’23, Red Ho…
  20. 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 EMNLP.
  21. Alberto Sanfeliu and King-Sun Fu. 1983. A distance measure between attributed relational graphs for pattern recognition. IEEE Transactions on Systems, Man, and Cybernetics, SMC-13(3):353–362. find
  22. Xiaoshuai Song, Keqing He, Pei Wang, Guanting Dong, Yutao Mou, Jingang Wang, Yunsen Xian, Xunliang Cai, and Weiran Xu. 2023a. Large language models meet open-world intent discovery and recognition: An evaluation of ChatGPT. In Proceedings of the 2023 Conference on Empirical Methods in Natural Langua… find
  23. Xiaoshuai Song, Yutao Mou, Keqing He, Yueyan Qiu, Jinxu Zhao, Pei Wang, and Weiran Xu. 2023b. Continual generalized intent discovery: Marching towards dynamic and open-world intent recognition. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 4370–4382, Singapore. Asso… find
  24. Jiayin Wang, Fengran Mo, Weizhi Ma, Peijie Sun, Min Zhang, and Jian-Yun Nie. 2024a. A user-centric multi-intent benchmark for evaluating large language models. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 3588–3612, Miami, Florida, USA. Association… find
  25. Lei Wang, Wanyu Xu, Yihuai Lan, Zhiqiang Hu, Yunshi Lan, Roy Ka-Wei Lee, and Ee-Peng Lim. 2023. Plan-and-solve prompting: Improving zero-shot chain-of-thought reasoning by large language models. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Lon… find
  26. Pei Wang, Keqing He, Yejie Wang, Xiaoshuai Song, Yutao Mou, Jingang Wang, Yunsen Xian, Xunliang Cai, and Weiran Xu. 2024b. Beyond the known: Investigating LLMs performance on out-of-domain intent detection. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Langu… find
  27. Chien-Sheng Wu, Andrea Madotto, Ehsan Hosseini-Asl, Caiming Xiong, Richard Socher, and Pascale Fung. 2019. Transferable multi-domain state generator for task-oriented dialogue systems. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 808–819, Florence…
  28. Zeqiu Wu, Yi Luan, Hannah Rashkin, David Reitter, Hannaneh Hajishirzi, Mari Ostendorf, and Gaurav Singh Tomar. 2022. CONQRR: Conversational query rewriting for retrieval with reinforcement learning. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 1000…
  29. Jian Xie, Kai Zhang, Jiangjie Chen, Tinghui Zhu, Renze Lou, Yuandong Tian, Yanghua Xiao, and Yu Su. 2024. Travelplanner: A benchmark for real-world planning with language agents. In Forty-first International Conference on Machine Learning. find
  30. Heng-Da Xu, Xian-Ling Mao, Puhai Yang, Fanshu Sun, and Heyan Huang. 2024a. 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–… find
  31. Weijie Xu, Zicheng Huang, Wenxiang Hu, Xi Fang, Rajesh Cherukuri, Naumaan Nayyar, Lorenzo Malandri, and Srinivasan Sengamedu. 2024b. HR-MultiWOZ: A task oriented dialogue (TOD) dataset for HR LLM agent. In Proceedings of the First Workshop on Natural Language Processing for Human Resources (NLP4HR 2… find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/mitra-2026-recap-rewriting-conversations-intent-understanding/