Elamur Research⌕
☰
home / papers / 2026 / STRIDE-ED: A Strategy-Grounded Stepwise Reasoning Framework for Empathetic Dialogue Systems

STRIDE-ED: A Strategy-Grounded Stepwise Reasoning Framework for Empathetic Dialogue Systems

Paper · 2026 · Hongru Ji, Yuyin Fan, Meng Zhao et al.

Summary

STRIDE-ED addresses a weakness in empathetic dialogue systems. Current models, including chain-of-thought LLM approaches, rarely ground their reasoning in a full strategy framework. Existing strategy sets come mostly from counseling and focus on negative emotions. The authors construct a broader empathy framework whose strategies span negative, neutral and positive emotional states alike. Each strategy gets a difficulty level (I–III), and the old Question strategy is split into three Exploring-type strategies. Responses are generated in explicit steps: a scenario summary, emotion recognition, strategy choice, concrete execution actions, and finally the reply. For training data, they use LLMs to annotate EMPATHETICDIALOGUES with strategies and rationales. The annotations are then filtered by multi-model consistency-weighted scoring and strategy-aware, difficulty-aware sampling. Training has two stages: supervised fine-tuning, then multi-objective reinforcement learning (PPO). The rewards target emotion, strategy adherence and output format. Across several open-source LLMs, the framework beats prior methods on automatic metrics and human ratings. This suggests that structured, strategy-grounded reasoning makes empathetic responses both more interpretable and more effective.

Key points

  • The full model scores BLEU-1 24.66, emotion accuracy 57.57, Dist-2 13.68 and PPL 9.26 on EMPATHETICDIALOGUES.
  • Ablations: removing any component hurts at least one metric. Dropping both data scoring/ranking and strategy-aware sampling causes the largest BLEU-1 loss (22.22). Removing PPO lowers emotion accuracy to 54.48.
  • Data scale (ED-CSA-all): performance holds up to 1/8 of the training data but collapses at 1/16 (BLEU-1 3.32).
  • Backbones tested: Qwen3-0.6B, Qwen3-4B, Qwen3-4B-Instruct, LLama3.2-3B and GLM-Z1-9B.
    • LLama3.2-3B is best on emotion accuracy (57.06) and diversity.
    • Qwen3-4B-Instruct is strongest on BLEU-1.
  • The strategy taxonomy extends the eight-strategy scheme of Liu et al. (2021).

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

🎓Northwestern Polytechnical UniversityXi'an · China13 works · 67 citations

Also on the byline: Henan University of Technology.

 Fields, topics and concepts

📚 Shelves 6The 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 worksExperimental and Cognitive Psychology · 870 works

🧭 Topics 3Research subjects and languages the work is about.Reasoning · 1,426 worksSentiment & emotion · 972 worksDialogue systems · 624 works

Cite

Cite as Hongru Ji, Yuyin Fan, Meng Zhao, Xianghua Li, Lianwei Wu and Chao Gao. 2026. STRIDE-ED: A Strategy-Grounded Stepwise Reasoning Framework for Empathetic Dialogue Systems. 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.321

@inproceedings{ji2026stride,
  title = {{STRIDE-ED: A Strategy-Grounded Stepwise Reasoning Framework for Empathetic Dialogue Systems}},
  author = {Hongru Ji, Yuyin Fan, Meng Zhao, Xianghua Li, Lianwei Wu and Chao Gao},
  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.321},
  url = {https://doi.org/10.18653/v1/2026.acl-long.321}
}

References

46 references, 17 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. C Daniel Batson. 2009. These things called empathy: Eight related but distinct phenomena. The Social Neuroscience of Empathy, pages 3–16. find
  2. Hua Cai, Xuli Shen, Qing Xu, Weilin Shen, Xiaomei Wang, Weifeng Ge, Xiaoqing Zheng, and Xiangyang Xue. 2023. Improving empathetic dialogue generation by dynamically infusing commonsense knowledge. In Findings of the Association for Computational Linguistics: ACL 2023, pages 7858–7873. find
  3. Jiawei Chen, Yang Yang, Chao Yu, Yu Tian, Zhi Cao, Xue Yang, Linghao Li, Hang Su, and Zhaoxia Yin. 2025a. Red teaming large reasoning models. arXiv preprint arXiv:2512.00412.
  4. Qi Chen and Dexi Liu. 2023. Dynamic strategy chain: Dynamic zero-shot cot for long mental health support generation. arXiv preprint arXiv:2308.10444. — [arXiv:2308.10444](https://arxiv.org/abs/2308.10444) find
  5. Yao Chen, Jiawei Sheng, Wenyuan Zhang, and Tingwen Liu. 2025b. Improving reasoning capabilities in small models through mixture-of-layers distillation with stepwise attention on key information. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 4952– 49…
  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, p… find
  7. LLMs are used for automatic annotation under prompts enforcing ethical constraints, and they are also applied to the grammatical checking and polishing of texts. Human evaluators are informed Fengyi Fu, Lei Zhang, Quan Wang, and Zhendong Mao. 2023. E-core: Emotion correlation enhanced empathetic dia… find
  8. Pan Gao, Donghong Han, Rui Zhou, Xuejiao Zhang, and Zikun Wang. 2023. Cab: empathetic dialogue generation with cognition, affection and behavior. In International Conference on Database Systems for Advanced Applications, pages 597–606. Springer. find
  9. Deepanway Ghosal, Navonil Majumder, Alexander Gelbukh, Rada Mihalcea, and Soujanya Poria. 2020. Cosmic: Commonsense knowledge for emotion identification in conversations. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 2470–2481. find
  10. Jiawang Hao and Fang Kong. 2025. Enhancing emotional support conversations: A framework for dynamic knowledge filtering and persona extraction. In Proceedings of the 31st International Conference on Computational Linguistics, pages 3193–3202. find
  11. He Hu, Yucheng Zhou, Juzheng Si, Qianning Wang, Hengheng Zhang, Fuji Ren, Fei Ma, and Laizhong Cui. 2025. Beyond empathy: Integrating diagnostic and therapeutic reasoning with large language models for mental health counseling. arXiv e-prints, pages arXiv–2505. find
  12. Hongru Ji, Xianghua Li, Mingxin Li, Meng Zhao, and Chao Gao. 2025. Hybrid relational graphs with sentiment-laden semantic alignment for multimodal emotion recognition in conversation. In Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, pages 2973–2981. find
  13. Liting Jiang, Di Wu, Yanbing Li, and Wushouer Silamu. 2023. Emp-usir: A unidirectional synchronous interactive reasoning model for empathetic dialogue. In 2023 International Joint Conference on Neural Networks (IJCNN), pages 1–9. IEEE. find
  14. Guocong Li, Jinjian Zhang, Ping Wang, Dongnan Liu, Tian Liang, Qiuyi Qi, Hao Huang, Siyan Guo, Mutian Bao, Wei Zhou, Linjian Mo, Hongxia Xu, and Jian Wu. 2026a. Mol: Adaptive mixture-of-length reasoning for efficient question answering with context. In The Fourteenth International Conference on Lear… find
  15. Junlin Li, Bo Peng, and Yu-Yin Hsu. 2024. Emstremo: Adapting emotional support response with enhanced emotion-strategy integrated selection. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 5794– 5805. find
  16. 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, pages 4454– 4466. find
  17. Qintong Li, Piji Li, Zhaochun Ren, Pengjie Ren, and Zhumin Chen. 2022. Knowledge bridging for empathetic dialogue generation. In Proceedings of the AAAI conference on artificial intelligence, volume 36, pages 10993–11001. find
  18. Yu Li, Rui Miao, Zhengling Qi, and Tian Lan. 2026b. Arise: Agent reasoning with intrinsic skill evolution in hierarchical reinforcement learning. arXiv preprint arXiv:2603.16060. — [arXiv:2603.16060](https://arxiv.org/abs/2603.16060) find
  19. 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), pages…
  20. Hongye Liu and Ricardo Henao. 2025. Learning to substitute words with model-based score ranking. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), pages 11551–11565. find
  21. 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…
  22. Tao Liu, Taiqiang Wu, Runming Yang, Shaoning Sun, Junjie Wang, and Yujiu Yang. 2026. Profit: Leveraging high-value signals in sft via probability-guided token selection. arXiv preprint arXiv:2601.09195.
  23. Ye Liu, Wolfgang Maier, Wolfgang Minker, and Stefan Ultes. 2022. Empathetic dialogue generation with pre-trained roberta-gpt2 and external knowledge. In Conversational AI for Natural Human-Centric Interaction: 12th International Workshop on Spoken Dialogue System Technology, IWSDS 2021, Singapore, p… find
  24. Guanran Luo, Wentao Qiu, Zhongquan Jian, Meihong Wang, and Qingqiang Wu. 2026. Gcot-decoding: Unlocking deep reasoning paths for universal question answering. Preprint, arXiv:2604.06794.
  25. Wei Peng, Yue Hu, Luxi Xing, Yuqiang Xie, Yajing Sun, and Yunpeng Li. 2022. Control globally, understand locally: A global-to-local hierarchical graph network for emotional support conversation. Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence, pages 4324–433… find
  26. Baiyou Qiao, Yuekai Zhang, Pan Gao, Xinchi Li, Shuo Wang, and Donghong Han. 2025. Multi-perspective empathy modeling for empathetic dialogue generation. Knowledge-Based Systems, 314:113191. find
  27. 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 annual meeting of the association for computational linguistics, pages 5370–5381. find
  28. Sahand Sabour, Chujie Zheng, and Minlie Huang. 2022. Cem: Commonsense-aware empathetic response generation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 36, pages 11229– 11237. find
  29. Lei Jiang, Yu Lu, Haibo Shi, Shuang Liang, Minlong Peng, and Flora D. Salim. 2026b. Why supervised fine-tuning fails to learn: A systematic study of incomplete learning in large language models. Preprint, arXiv:2604.10079.
  30. Maarten Sap, Ronan Le Bras, Emily Allaway, Chandra Bhagavatula, Nicholas Lourie, Hannah Rashkin, Brendan Roof, Noah A Smith, and Yejin Choi. 2019. Atomic: An atlas of machine commonsense for if-then reasoning. In Proceedings of the AAAI conference on artificial intelligence, volume 33, pages 3027–30… find
  31. Zhou Yang, Zhaochun Ren, Wang Yufeng, Haizhou Sun, Chao Chen, Xiaofei Zhu, and Xiangwen Liao. 2024. An iterative associative memory model for empathetic response generation. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3081… find
  32. C Spearman. 1904. The proof and measurement of association between two things. The American Journal of Psychology, 15(1):72–101. find
  33. Jing Ye, Lu Xiang, Yaping Zhang, and Chengqing Zong. 2025. Sweetiechat: A strategy-enhanced role-playing framework for diverse scenarios handling emotional support agent. In Proceedings of the 31st International Conference on Computational Linguistics, pages 4646–4669.
  34. Quan Tu, Yanran Li, Jianwei Cui, Bin Wang, Ji-Rong Wen, and Rui Yan. 2022. Misc: A mixed strategy-aware model integrating comet for emotional support conversation. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 308–319.
  35. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. Advances in neural information processing systems, 30.
  36. Lanrui Wang, Jiangnan Li, Chenxu Yang, Zheng Lin, Hongyin Tang, Huan Liu, Yanan Cao, Jingang Wang, and Weiping Wang. 2025. Sibyl: Empowering empathetic dialogue generation in large language models via sensible and visionary commonsense inference. In Proceedings of the 31st International Conference o…
  37. Yu Wang, Emmanuele Chersoni, and Chu-Ren Huang. 2026. This one or that one? a study on accessibility via demonstratives with multimodal large language models. In Language Resources and Evaluation Conference 2026. European Language Resources Association (ELRA).
  38. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed Chi, Quoc Le, and Denny Zhou. 2022. Chain-of-thought prompting elicits reasoning in large language models. Advances in neural information processing systems, 35:24824– 24837.
  39. Chao Xue, Yao Wang, Mengqiao Liu, Di Liang, Xingsheng Han, Peiyang Liu, Xianjie Wu, Chenyao Lu, Lei Jiang, Yu Lu, Haibo Shi, Shuang Liang, Minlong Peng, and Flora D. Salim. 2026a. Reason only when needed: Efficient generative reward modeling via model-internal uncertainty. Preprint, arXiv:2604.10072…
  40. Chao Xue, Yao Wang, Mengqiao Liu, Di Liang, Xingsheng Han, Peiyang Liu, Xianjie Wu, Chenyao Lu, Jiahao Yuan, Zixiang Di, Zhiqing Cui, Guisong Yang, and Usman Naseem. 2025. Reflectdiffu: Reflect between emotion-intent contagion and mimicry for empathetic response generation via a rl-diffusion framewo… find
  41. Qianchi Zhang, Hainan Zhang, Liang Pang, Hongwei Zheng, and Zhiming Zheng. 2026a. Stable-rag: Mitigating retrieval-permutation-induced hallucinations in retrieval-augmented generation. arXiv preprint arXiv:2601.02993.
  42. Tenggan Zhang, Xinjie Zhang, Jinming Zhao, Li Zhou, and Qin Jin. 2024. Escot: Towards interpretable emotional support dialogue systems. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 13395–13412. find
  43. Xinglang Zhang, Yunyao Zhang, ZeLiang Chen, Junqing Yu, Wei Yang, and Zikai Song. 2026b. Logical phase transitions: Understanding collapse in llm logical reasoning. Preprint, arXiv:2601.02902.
  44. Xinjie Zhang, Wenxuan Wang, and Qin Jin. 2025. In-tentionesc: An intention-centered framework for enhancing emotional support in dialogue systems. In Findings of the Association for Computational Linguistics: ACL 2025, pages 26494–26516, Vienna, Austria. Association for Computational Linguistics. find
  45. Yunyao Zhang, Xinglang Zhang, Junxi Sheng, Wenbing Li, Junqing Yu, Yi-Ping Phoebe Chen, Wei Yang, and Zikai Song. 2026c. Semantic-aware logical reasoning via a semiotic framework. Preprint, arXiv:2509.24765.
  46. Peixiang Zhong, Di Wang, Pengfei Li, Chen Zhang, Hao Wang, and Chunyan Miao. 2021. Care: commonsense-aware emotional response generation with latent concepts. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, pages 14577–14585. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/ji-2026-stride-ed-strategy-grounded-stepwise/