Elamur Research⌕
☰
home / papers / 2026 / Everyone is unique: Towards Behaviorally Heterogeneous Negotiation Dialogue Systems for Debt Collection

Everyone is unique: Towards Behaviorally Heterogeneous Negotiation Dialogue Systems for Debt Collection

Paper · 2026 · Yuhang Yang, Kai Tang, Chao Ye et al.

Summary

Debt collection is a demanding negotiation setting. Existing negotiation benchmarks treat users as rational agents whose preferences never change, but real debtors vary widely in emotion, reasoning ability and manner of speaking. The authors had access to confidential collector–debtor conversations from a fintech partner. Because those conversations cannot be published, they turned the behavioral patterns into a synthetic, privacy-preserving benchmark called DebtBench. It is built in three stages: personas are extracted with an LLM, collector and debtor strategy sets are derived through clustering and expert review, and an LLM judge iteratively refines utterances so they match each persona. The authors also train DebtGPT with Coarse-to-Fine Preference Optimization (CFPO), which balances long-term repayment against the debtor's interaction experience. Across 16 LLMs, most systems perform poorly, and reasoning-oriented models do worse than general-purpose ones. The 8B DebtGPT beats every open-source baseline and is roughly on par with GPT-4o. The work offers an ethically constructed testbed for empathetic, behavior-aware negotiation agents in sensitive domains.

Key points

  • Personas come from 1,000 real conversations and have four dimensions: background, personality (including MBTI and linguistic style), cognitive attributes (legal and financial understanding) and a life-grounded scenario.
  • Strategies were clustered with HDBSCAN, giving 9 collector strategies and 8 debtor strategies, all validated by domain experts.
  • Most of the 16 evaluated LLMs reach success rates below 75%, often conceding too much or failing to adapt to the debtor.
  • DebtGPT (8B, trained with CFPO) outperforms all open-source baselines and performs comparably to GPT-4o.
  • In a human study with 20 finance professionals, about 73% of LLM-judge interaction scores fell within ±2.15 of human scores. Code and data are at github.com/YYuHhhh/DebtNegotiation under a research-only license.

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 · 18,614 worksArtificial Intelligence · 17,965 worksArts and Humanities · 17,450 worksLanguage and Linguistics · 17,296 works

🧭 Topics 1Research subjects and languages the work is about.Dialogue systems · 654 works

Cite

Cite as Yuhang Yang, Kai Tang, Chao Ye, Haobo Wang, Qiqi Luo, Jin Guang Zheng and Zhixin Zhang. 2026. Everyone is unique: Towards Behaviorally Heterogeneous Negotiation Dialogue Systems for Debt Collection. 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.232

@inproceedings{yang2026everyone,
  title = {{Everyone is unique: Towards Behaviorally Heterogeneous Negotiation Dialogue Systems for Debt Collection}},
  author = {Yuhang Yang, Kai Tang, Chao Ye, Haobo Wang, Qiqi Luo, Jin Guang Zheng and Zhixin Zhang},
  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.232},
  url = {https://doi.org/10.18653/v1/2026.acl-long.232}
}

References

42 references, 20 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Naoki Abe, Prem Melville, Cezar Pendus, Chandan K Reddy, David L Jensen, Vince P Thomas, James J Bennett, Gary F Anderson, Brent R Cooley, Melissa Kowalczyk, and 1 others. 2010. Optimizing debt collections using constrained reinforcement learning. In Proceedings of the 16th ACM SIGKDD international… find
  2. Anthropic. 2025. Claude sonnet 4.0. https://claude.ai. Large language model. ↗
  3. Carol Byrne. 2024. The understated dangers of poor customer communications in collections. Accessed: 2024-06-15. find
  4. Kushal Chawla, Jaysa Ramirez, Rene Clever, Gale Lucas, Jonathan May, and Jonathan Gratch. 2021. Casino: A corpus of campsite negotiation dialogues for automatic negotiation systems. arXiv preprint arXiv:2103.15721. find
  5. Gheorghe Comanici, Eric Bieber, Mike Schaekermann, Ice Pasupat, Noveen Sachdeva, Inderjit S. Dhillon, Marcel Blistein, Ori Ram, Dan Zhang, Evan Rosen, Luke Marris, Sam Petulla, Colin Gaffney, Asaf Aharoni, Nathan Lintz, Tiago Cardal Pais, Henrik Jacobsson, Idan Szpektor, Nan-Jiang Jiang, and 81 othe… — Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
  6. DeepSeek-AI. 2025. Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning. Preprint, arXiv:2501.12948.
  7. Yang Deng, Lizi Liao, Liang Chen, Hongru Wang, Wenqiang Lei, and Tat-Seng Chua. 2023. Prompting and evaluating large language models for proactive dialogues: Clarification, target-guided, and non-collaboration. In Findings of the Association for Computational Linguistics: EMNLP 2023, Singapore, Dece…
  8. Yang Deng, Wenxuan Zhang, Wai Lam, See-Kiong Ng, and Tat-Seng Chua. 2024. Plug-and-play policy planner for large language model powered dialogue agents. In The Twelfth International Conference on Learning Representations, ICLR 2024, Vienna, Austria, May 7-11, 2024. OpenReview.net. find
  9. Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, and 1 others. 2024. The llama 3 herd of models. arXiv preprint arXiv:2407.21783.
  10. Gika Firanda, Paramita Prananingtyas, and Sartika Lestari. 2021. Debt collection of financial technology lending. In 1st International Conference on Science and Technology in Administration and Management Information. find
  11. Joseph L Fleiss. 1971. Measuring nominal scale agreement among many raters. Psychological bulletin, 76(5):378. find
  12. Yao Fu, Hao Peng, Tushar Khot, and Mirella Lapata. 2023. Improving language model negotiation with self-play and in-context learning from AI feedback. CoRR, abs/2305.10142.
  13. He He, Derek Chen, Anusha Balakrishnan, and Percy Liang. 2018. Decoupling strategy and generation in negotiation dialogues. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, Brussels, Belgium, October 31 -November 4, 2018, pages 2333–2343. Association for Com…
  14. Tao He, Lizi Liao, Yixin Cao, Yuanxing Liu, Ming Liu, Zerui Chen, and Bing Qin. 2024. Planning like human: A dual-process framework for dialogue planning. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2024, Bangkok, Thailand,… find
  15. Edward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. 2022. Lora: Low-rank adaptation of large language models. In The Tenth International Conference on Learning Representations, ICLR 2022, Virtual Event, April 25-29, 2022. OpenReview.net.
  16. Chuhao Jin, Kening Ren, Lingzhen Kong, Xiting Wang, Ruihua Song, and Huan Chen. 2024. Persuading across diverse domains: a dataset and persuasion large language model. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1678– 1706… find
  17. Dexin Kong, Xu Yan, Ming Chen, Shuguang Han, Jufeng Chen, and Fei Huang. 2025. Fishbargain: An llm-empowered bargaining agent for online flea-market platform sellers. In Companion Proceedings of the ACM on Web Conference 2025, WWW 2025, Sydney, NSW, Australia, 28 April 2025 - 2 May 2025, pages 2855–… find
  18. Woosuk Kwon, Zhuohan Li, Siyuan Zhuang, Ying Sheng, Lianmin Zheng, Cody Hao Yu, Joseph Gonzalez, Hao Zhang, and Ion Stoica. 2023. Efficient memory management for large language model serving with pagedattention. In Proceedings of the 29th Symposium on Operating Systems Principles, SOSP 2023, Koblenz…
  19. Mike Lewis, Denis Yarats, Yann N Dauphin, Devi Parikh, and Dhruv Batra. 2017. Deal or no deal? end-to-end learning for negotiation dialogues. arXiv preprint arXiv:1706.05125. — [arXiv:1706.05125](https://arxiv.org/abs/1706.05125) find
  20. Weixin Liang, Mert Yuksekgonul, Yining Mao, Eric Wu, and James Zou. 2023. Gpt detectors are biased against non-native english writers. Patterns, 4(7).
  21. Aixin Liu, Bei Feng, Bing Xue, Bingxuan Wang, Bochao Wu, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenyu Zhang, Chong Ruan, and 1 others. 2024. Deepseek-v3 technical report. arXiv preprint arXiv:2412.19437. find
  22. Yang Liu, Dan Iter, Yichong Xu, Shuohang Wang, Ruochen Xu, and Chenguang Zhu. 2023. G-eval: Nlg evaluation using gpt-4 with better human alignment. arXiv preprint arXiv:2303.16634.
  23. Mary L McHugh. 2012. Interrater reliability: the kappa statistic. Biochemia medica, 22(3):276–282. find
  24. Leland McInnes, John Healy, Steve Astels, and 1 others. 2017. hdbscan: Hierarchical density based clustering. J. Open Source Softw., 2(11):205.
  25. OpenAI. 2023. GPT-4 technical report. abs/2303.08774.
  26. CoRR, Peterson K Ozili. 2019. Non-performing loans and financial development: new evidence. The Journal of Risk Finance, 20(1):59–81. find
  27. Rafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D. Manning, Stefano Ermon, and Chelsea Finn. 2023. Direct preference optimization: Your language model is secretly a reward model. In Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Sy…
  28. Garden Tabacchi, Claudio Costantino, Giuseppe Napoli, Valentina Marchese, Manuela Cracchiolo, Alessandra Casuccio, Francesco Vitale, Esculapio Working Group, and 1 others. 2016. Determinants of european parents’ decision on the vaccination of their children against measles, mumps and rubella: A syst… find
  29. Kimi Team, Yifan Bai, Yiping Bao, Guanduo Chen, Jiahao Chen, Ningxin Chen, Ruijue Chen, Yanru Chen, Yuankun Chen, Yutian Chen, and 1 others. 2025. Kimi k2: Open agentic intelligence. arXiv preprint arXiv:2507.20534.
  30. Issue Yishu Wang, Kakam Chong, Xiaofeng Wang, Xu Yan, DeXin Kong, Chen Ju, Ming Chen, Shuai Xiao, Shuguang Han, and 1 others. 2025a. Evaluating multi-turn bargain skills in llm-based seller agent. arXiv preprint arXiv:2509.06341. — [arXiv:2509.06341](https://arxiv.org/abs/2509.06341) find
  31. Peiyi Wang, Lei Li, Liang Chen, Zefan Cai, Dawei Zhu, Binghuai Lin, Yunbo Cao, Lingpeng Kong, Qi Liu, Tianyu Liu, and 1 others. 2024. Large language models are not fair evaluators. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pag… find
  32. Xiaofeng Wang, Zhixin Zhang, Jinguang Zheng, Yiming Ai, and Rui Wang. 2025b. Debt collection negotiations with large language models: An evaluation system and optimizing decision making with multi-agent. arXiv preprint arXiv:2502.18228. find
  33. Xuewei Wang, Weiyan Shi, Richard Kim, Yoojung Oh, Sijia Yang, Jingwen Zhang, and Zhou Yu. 2019. Persuasion for good: Towards a personalized persuasive dialogue system for social good. arXiv preprint arXiv:1906.06725.
  34. Atsuki Yamaguchi, Kosui Iwasa, and Katsuhide Fujita. 2021. Dialogue act-based breakdown detection in negotiation dialogues. In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, pages 745–757.
  35. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, Chujie Zheng, Dayiheng Liu, Fan Zhou, Fei Huang, Feng Hu, Hao Ge, Haoran Wei, Huan Lin, Jialong Tang, and 40 others. 2025. Qwen3 technical report. CoRR, abs/2505.09388.
  36. An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, and 1 others. 2024. Qwen2. 5 technical report. arXiv preprint arXiv:2412.15115. find
  37. Tracy Yang, Tian Lu, Beibei Li, and Lu Xianghua. 2020. Personalizing debt collections: Combining reinforcement learning and field experiment. find
  38. Xiao Yu, Maximillian Chen, and Zhou Yu. 2023. Prompt-based monte-carlo tree search for goal-oriented dialogue policy planning. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, EMNLP 2023, Singapore, December 6-10, 2023, pages 7101–7125. Association for Compu… find
  39. Aohan Zeng, Xin Lv, Qinkai Zheng, Zhenyu Hou, Bin Chen, Chengxing Xie, Cunxiang Wang, Da Yin, Hao Zeng, Jiajie Zhang, and 1 others. 2025. Glm-4.5: Agentic, reasoning, and coding (arc) foundation models. arXiv preprint arXiv:2508.06471.
  40. Qiang Zhang, Jason Naradowsky, and Yusuke Miyao. 2023. Ask an expert: Leveraging language models to improve strategic reasoning in goal-oriented dialogue models. In Findings of the Association for Computational Linguistics: ACL 2023, Toronto, Canada, July 9-14, 2023, pages 6665–6694. Association for… find
  41. Yaowei Zheng, Richong Zhang, Junhao Zhang, Yanhan Ye, Zheyan Luo, and Yongqiang Ma. 2024. Llamafactory: Unified efficient fine-tuning of 100+ language models. CoRR, abs/2403.13372.
  42. Xuhui Zhou, Hao Zhu, Leena Mathur, Ruohong Zhang, Haofei Yu, Zhengyang Qi, Louis-Philippe Morency, Yonatan Bisk, Daniel Fried, Graham Neubig, and Maarten Sap. 2024. SOTOPIA: interactive evaluation for social intelligence in language agents. In The Twelfth International Conference on Learning Represe…

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/yang-2026-everyone-is-unique-behaviorally-heterogeneous/