Elamur Research⌕
☰
home / papers / 2025 / Crisp: Cognitive Restructuring of Negative Thoughts through Multi-turn Supportive Dialogues

Crisp: Cognitive Restructuring of Negative Thoughts through Multi-turn Supportive Dialogues

Paper · 2025 · Jinfeng Zhou, Yuxuan Chen, Jianing Yin et al.

Summary

Cognitive Restructuring (CR) is a therapy technique in which a counselor helps someone uncover distorted negative thoughts and reshape them into more balanced ones through conversation. Therapists are scarce and seeking help carries stigma, so LLM-based CR is appealing. However, earlier systems either rewrite thoughts directly, follow rigid short scripts, or handle only one distortion per session. These approaches ignore emotional resistance and individual variation. The authors introduce CRDial, a framework that splits CR into two theory-based stages. First, Cognitive Theory guides the uncovering of thoughts, moving from surface to core beliefs. In the second stage, those thoughts are challenged by applying the Defense Attorney Technique. CRDial also adds sentence-level supportive strategies and a loop that lets several distortions be addressed within one conversation. Using CRDial with GPT-4o self-play, they build Crisp, a bilingual corpus of about 22k dialogues that passed quality and safety checks. They then fine-tune Crispers models at 7B and 14B. Human evaluations and an intervention study show Crispers-14B surpassing its GPT-4o teacher. This suggests that compact open models can support research on interactive, LLM-assisted psychotherapy.

Key points

  • CRDial framework has three parts:
    • a two-stage dialogue: CT-guided identification of automatic thoughts, intermediate beliefs and core distortions, followed by DAT-driven "defense/prosecution" restructuring;
    • a supportive-strategy scheme with 5 categories and 8 sub-categories (e.g., Reflecting Feelings, Restatement, Affirmation and Reassurance);
    • a multi-channel loop for handling multiple distortions.
  • Crisp contains roughly 22k bilingual multi-stage, multi-turn dialogues distilled from GPT-4o, with filtering for quality, safety and professionalism. Code and data are at github.com/thu-coai/Crisp.
  • Crispers are conversational LLMs trained on Crisp at 7B and 14B scales. In interactive pointwise and pairwise human evaluations, Crispers-14B outperformed its GPT-4o teacher.
  • Intervention trial: 90 participants were chosen from 200 applicants using PANAS, then split into three groups (initial scores did not differ, ANOVA p = 0.74).
    • Within its group, Crispers-14B increased positive affect by 49% and reduced negative affect by 44% (p<0.001).
    • It significantly beat Emohaa (p<0.01) and prompted GPT-4o (p<0.05) under Tukey HSD.
  • Stated limitations: reliance on a single generator LLM, possible annotator bias, and no claims of effectiveness for high-risk cases such as self-harm. The study received IRB approval, and the resources are released for research only.

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 PennsylvaniaPhiladelphia · United States221 works · 63,234 citations🎓Tsinghua UniversityBeijing · China333 works · 87,019 citations🎓Harvard UniversityCambridge · United States112 works · 42,023 citations🎓Zhejiang UniversityHangzhou · China172 works · 12,988 citations

 Fields, topics and concepts

📚 Shelves 7The 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 worksPsychology · 1,552 worksSocial Psychology · 306 worksApplied Psychology · 21 works

🧭 Topics 1Research subjects and languages the work is about.Dialogue systems · 624 works

Cite

Cite as Jinfeng Zhou, Yuxuan Chen, Jianing Yin, Yongkang Huang, Yihan Shi, Xikun Zhang, Libiao Peng, Rongsheng Zhang, Tangjie Lv, Zhipeng Hu, Hongning Wang and Minlie Huang. 2025. Crisp: Cognitive Restructuring of Negative Thoughts through Multi-turn Supportive Dialogues. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2025.emnlp-main.1652

@inproceedings{zhou2025crisp,
  title = {{Crisp: Cognitive Restructuring of Negative Thoughts through Multi-turn Supportive Dialogues}},
  author = {Jinfeng Zhou, Yuxuan Chen, Jianing Yin, Yongkang Huang, Yihan Shi, Xikun Zhang, Libiao Peng, Rongsheng Zhang, Tangjie Lv, Zhipeng Hu, Hongning Wang and Minlie Huang},
  booktitle = {Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing},
  year = {2025},
  doi = {10.18653/v1/2025.emnlp-main.1652},
  url = {https://doi.org/10.18653/v1/2025.emnlp-main.1652}
}

References

50 references, 11 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Navneet Agarwal and Kairit Sirts. 2025. Exploratory study into relations between cognitive distortions and emotional appraisals. Preprint, arXiv:2503.15979. find
  2. American Psychological Association. 2017. Ethical principles of psychologists and code of conduct. Amended effective June 1, 2010, and January 1, 2017. find
  3. John A Bargh and Tanya L Chartrand. 1999. The unbearable automaticity of being. American psychologist, 54(7):462. find
  4. Aaron T Beck. 1979. Cognitive therapy and the emotional disorders. Penguin. find
  5. Judith S Beck. 2011. Cognitive behavior therapy: Basics and beyond. Guilford Publications. find
  6. Yirong Chen, Xiaofen Xing, Jingkai Lin, Huimin Zheng, Zhenyu Wang, Qi Liu, and Xiangmin Xu. 2023. Soulchat: Improving llms’ empathy, listening, and comfort abilities through fine-tuning with multi-turn empathy conversations. In Findings of the Association for Computational Linguistics: EMNLP 2023, S… find
  7. Irismar Reis de Oliveira. 2011. Kafka’s trial dilemma: proposal of a practical solution to joseph k.’s unknown accusation. Medical Hypotheses, 77(1):5–6. find
  8. Ronald Aylmer Fisher. 1970. Statistical methods for research workers. In Breakthroughs in statistics: Methodology and distribution, pages 66–70. Springer. find
  9. Joseph L Fleiss. 1971. Measuring nominal scale agreement among many raters. Psychological bulletin, 76(5):378. find
  10. Team GLM, :, Aohan Zeng, Bin Xu, Bowen Wang, Chenhui Zhang, Da Yin, Diego Rojas, Guanyu Feng, Hanlin Zhao, Hanyu Lai, Hao Yu, Hongning Wang, Jiadai Sun, Jiajie Zhang, Jiale Cheng, Jiayi Gui, Jie Tang, Jing Zhang, Juanzi Li, Lei Zhao, Lindong Wu, Lucen Zhong, Mingdao Liu, Minlie Huang, Peng Zhang, Qi… — ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools
  11. Anmol Goel, Nico Daheim, and Iryna Gurevych. 2024. Socratic reasoning improves positive text rewriting. CoRR, abs/2403.03029. find
  12. Clara E Hill. 2009. Helping skills: Facilitating exploration, insight, and action. American Psychological Association. find
  13. Jena D. Hwang, Chandra Bhagavatula, Ronan Le Bras, Jeff Da, Keisuke Sakaguchi, Antoine Bosselut, and Yejin Choi. 2021. (comet-) atomic 2020: On symbolic and neural commonsense knowledge graphs. In Thirty-Fifth AAAI Conference on Artificial Intelligence, AAAI 2021, Thirty-Third Conference on Innovati…
  14. Hyunwoo Kim, Jack Hessel, Liwei Jiang, Peter West, Ximing Lu, Youngjae Yu, Pei Zhou, Ronan Le Bras, Malihe Alikhani, Gunhee Kim, Maarten Sap, and Yejin Choi. 2023. SODA: million-scale dialogue distillation with social commonsense contextualization. In Proceedings of the 2023 Conference on Empirical…
  15. Hyunwoo Kim, Youngjae Yu, Liwei Jiang, Ximing Lu, Daniel Khashabi, Gunhee Kim, Yejin Choi, and Maarten Sap. 2022. Prosocialdialog: A prosocial backbone for conversational agents. Preprint, arXiv:2205.12688.
  16. Subin Kim, Hoonrae Kim, Heejin Do, and Gary Geunbae Lee. 2025a. Multimodal cognitive reframing therapy via multi-hop psychotherapeutic reasoning. CoRR, abs/2502.06873. find
  17. Subin Kim, Hoonrae Kim, Heejin Do, and Gary Geunbae Lee. 2025b. Multimodal cognitive reframing therapy via multi-hop psychotherapeutic reasoning. Preprint, arXiv:2502.06873. find
  18. Subin Kim, Hoonrae Kim, Jihyun Lee, Yejin Jeon, and Gary Geunbae Lee. 2025c. Mirror: Multimodal cognitive reframing therapy for rolling with resistance. Preprint, arXiv:2504.13211. find
  19. Suyeon Lee, Sunghwan Kim, Minju Kim, Dongjin Kang, Dongil Yang, Harim Kim, Minseok Kang, Dayi Jung, Min Hee Kim, Seungbeen Lee, Kyoung-Mee Chung, Youngjae Yu, Dongha Lee, and Jinyoung Yeo. 2024. Cactus: Towards psychological counseling conversations using cognitive behavioral theory. In Findings of… find
  20. Qintong Li, Hongshen Chen, Zhaochun Ren, Pengjie Ren, Zhaopeng Tu, and Zhumin Chen. 2020. Empdg: Multi-resolution interactive empathetic dialogue generation. In Proceedings of the 28th International Conference on Computational Linguistics, COLING 2020, Barcelona, Spain (Online), December 8-13, 2020,… find
  21. Zhaojiang Lin, Andrea Madotto, Jamin Shin, Peng Xu, and Pascale Fung. 2019. Moel: Mixture of empathetic listeners. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, EMNLP-IJCNLP 2019, H…
  22. Marsha Linehan. 2014. DBT? Skills training manual. Guilford Publications. find
  23. Siyang Liu, Chujie Zheng, Orianna Demasi, Sahand Sabour, Yu Li, Zhou Yu, Yong Jiang, and Minlie Huang. 2021. Towards emotional support dialog systems. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Lan…
  24. Ilya Loshchilov and Frank Hutter. 2019. Decoupled weight decay regularization. In 7th International Conference on Learning Representations, ICLR 2019, New Orleans, LA, USA, May 6-9, 2019. OpenReview.net. find
  25. Ryan Louie, Ananjan Nandi, William Fang, Cheng Chang, Emma Brunskill, and Diyi Yang. 2024. Roleplay-doh: Enabling domain-experts to create llm-simulated patients via eliciting and adhering to principles. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, EMNLP…
  26. Mounica Maddela, Megan Ung, Jing Xu, Andrea Madotto, Heather Foran, and Y-Lan Boureau. 2023. Training models to generate, recognize, and reframe unhelpful thoughts. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2023, Toronto,… find
  27. Philip M McCarthy and Scott Jarvis. 2010. Mtld, vocd-d, and hd-d: A validation study of sophisticated approaches to lexical diversity assessment. Behavior research methods, 42(2):381–392. find
  28. Sachin Mehta, Mohammad Hossein Sekhavat, Qingqing Cao, Maxwell Horton, Yanzi Jin, Chenfan Sun, Seyed-Iman Mirzadeh, Mahyar Najibi, Dmitry Belenko, Peter Zatloukal, and Mohammad Rastegari. 2024. Openelm: An efficient language model family with open training and inference framework. CoRR, abs/2404.146… — [arXiv:2404.14619](https://arxiv.org/abs/2404.14619) find
  29. Hongbin Na. 2024. CBT-LLM: A chinese large language model for cognitive behavioral therapy-based mental health question answering. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation, LREC/COLING 2024, 20-25 May, 2024, Torino, Ita… find
  30. Mark Olfson. 2016. Building the mental health workforce capacity needed to treat adults with serious mental illnesses. Health Affairs, 35(6):983–990. find
  31. OpenAI. 2024. Hello gpt-4o. find
  32. Yilin Qi, Dong Won Lee, Cynthia Breazeal, and Hae Won Park. 2025. Does "reasoning" with large language models improve recognizing, generating, and reframing unhelpful thoughts? Preprint, arXiv:2504.00163. find
  33. Huachuan Qiu, Hongliang He, Shuai Zhang, Anqi Li, and Zhenzhong Lan. 2024a. SMILE: single-turn to multi-turn inclusive language expansion via chatgpt for mental health support. In Findings of the Association for Computational Linguistics: EMNLP 2024, Miami, Florida, USA, November 12-16, 2024, pages…
  34. Huachuan Qiu and Zhenzhong Lan. 2024. Interactive agents: Simulating counselor-client psychological counseling via role-playing llm-to-llm interactions. CoRR, abs/2408.15787.
  35. Huachuan Qiu, Anqi Li, Lizhi Ma, and Zhenzhong Lan. 2024b. Psychat: A client-centric dialogue system for mental health support. Preprint, arXiv:2312.04262. — [arXiv:2312.04262](https://arxiv.org/abs/2312.04262) find
  36. Qwen, :, An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, Huan Lin, Jian Yang, Jianhong Tu, Jianwei Zhang, Jianxin Yang, Jiaxi Yang, Jingren Zhou, Junyang Lin, Kai Dang, Keming Lu, Keqin Bao, Kexin Yang, Le Yu, Mei Li, Mingfeng… find
  37. Hannah Rashkin, Eric Michael Smith, Margaret Li, and Y-Lan Boureau. 2019. Towards empathetic open-domain conversation models: A new benchmark and dataset. In Proceedings of the 57th Conference of the Association for Computational Linguistics, ACL 2019, Florence, Italy, July 28- August 2, 2019, Volum… find
  38. Ashish Sharma, Kevin Rushton, Inna E. Lin, David Wadden, Khendra G. Lucas, Adam S. Miner, Theresa Nguyen, and Tim Althoff. 2023. Cognitive reframing of negative thoughts through human-language model interaction. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguisti… find
  39. Len Sperry and Jon Sperry. 2020. Case conceptualization: Mastering this competency with ease and confidence. Routledge. find
  40. Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, Aurélien Rodriguez, Armand Joulin, Edouard Grave, and Guillaume Lample. 2023a. Llama: Open and efficient foundation language models. CoRR, a…
  41. Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, Dan Bikel, Lukas Blecher, Cristian Canton-Ferrer, Moya Chen, Guillem Cucurull, David Esiobu, Jude Fernandes, Jeremy Fu, Wenyin Fu, Brian Fuller,… — Llama 2: Open Foundation and Fine-Tuned Chat Models
  42. John W Tukey. 1949. Comparing individual means in the analysis of variance. Biometrics, pages 99–114. and negative affect: The panas scales. Journal of Personality and Social Psychology, 54(6):1063–1070. find
  43. Mengxi Xiao, Qianqian Xie, Ziyan Kuang, Zhicheng Liu, Kailai Yang, Min Peng, Weiguang Han, and Jimin Huang. 2024. Healme: Harnessing cognitive reframing in large language models for psychotherapy. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: L… find
  44. Chenhao Zhang, Renhao Li, Minghuan Tan, Min Yang, Jingwei Zhu, Di Yang, Jiahao Zhao, Guancheng Ye, Chengming Li, and Xiping Hu. 2024a. Cpsycoun: A report-based multi-turn dialogue reconstruction and evaluation framework for chinese psychological counseling. In Findings of the Association for Computa… find
  45. Tenggan Zhang, Xinjie Zhang, Jinming Zhao, Li Zhou, and Qin Jin. 2024b. Escot: Towards interpretable emotional support dialogue systems. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2024, Bangkok, Thailand, August 11-16, 2024… find
  46. Chujie Zheng, Sahand Sabour, Jiaxin Wen, Zheng Zhang, and Minlie Huang. 2023a. Augesc: Dialogue augmentation with large language models for emotional support conversation. In Findings of the Association for Computational Linguistics: ACL 2023, Toronto, Canada, July 9-14, 2023, pages 1552–1568. Assoc… find
  47. Zhonghua Zheng, Lizi Liao, Yang Deng, and Liqiang Nie. 2023b. Building emotional support chatbots in the era of llms. CoRR, abs/2308.11584. — [arXiv:2308.11584](https://arxiv.org/abs/2308.11584) find
  48. Jinfeng Zhou, Chujie Zheng, Bo Wang, Zheng Zhang, and Minlie Huang. 2023. CASE: aligning coarse-to-fine cognition and affection for empathetic response generation. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2023, Toronto, C… find
  49. Caleb Ziems, Minzhi Li, Anthony Zhang, and Diyi Yang. 2022. Inducing positive perspectives with text reframing. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), ACL 2022, Dublin, Ireland, May 22-27, 2022, pages 3682–3700. Association… find
  50. D. Watson, L. A. Clark, and A. Tellegen. 1988. Development and validation of brief measures of positive find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/zhou-2025-crisp-cognitive-restructuring-negative-thoughts/