TY - GEN
T1 - STRIDE-ED
T2 - 64th Annual Meeting of the Association for Computational Linguistics, ACL 2026
AU - Ji, Hongru
AU - Fan, Yuyin
AU - Zhao, Meng
AU - Li, Xianghua
AU - Wu, Lianwei
AU - Gao, Chao
N1 - Publisher Copyright:
© 2026 Association for Computational Linguistics.
PY - 2026
Y1 - 2026
N2 - Empathetic dialogue requires not only recognizing a user's emotional state but also making strategy-aware, context-sensitive decisions throughout response generation. However, the lack of a comprehensive empathy strategy framework, explicit task-aligned multi-stage reasoning, and high-quality strategy-aware data fundamentally limits existing approaches, preventing them from effectively modeling empathetic dialogue as a complex, multi-stage cognitive and decision-making process. To address these challenges, we propose STRIDE-ED, a STRategy-grounded, Interpretable, and DEep reasoning framework that models Empathetic Dialogue through structured, strategy-conditioned reasoning. To support effective learning, we develop a strategy-aware data refinement pipeline integrating LLM-based annotation, multi-model consistency-weighted evaluation, and dynamic sampling to construct high-quality training data aligned with empathetic strategies. Furthermore, we adopt a two-stage training paradigm that combines supervised fine-tuning with multi-objective reinforcement learning to better align model behaviors with target emotions, empathetic strategies, and response formats. Extensive experiments demonstrate that STRIDE-ED generalizes across diverse open-source LLMs and consistently outperforms existing methods on both automatic metrics and human evaluations.
AB - Empathetic dialogue requires not only recognizing a user's emotional state but also making strategy-aware, context-sensitive decisions throughout response generation. However, the lack of a comprehensive empathy strategy framework, explicit task-aligned multi-stage reasoning, and high-quality strategy-aware data fundamentally limits existing approaches, preventing them from effectively modeling empathetic dialogue as a complex, multi-stage cognitive and decision-making process. To address these challenges, we propose STRIDE-ED, a STRategy-grounded, Interpretable, and DEep reasoning framework that models Empathetic Dialogue through structured, strategy-conditioned reasoning. To support effective learning, we develop a strategy-aware data refinement pipeline integrating LLM-based annotation, multi-model consistency-weighted evaluation, and dynamic sampling to construct high-quality training data aligned with empathetic strategies. Furthermore, we adopt a two-stage training paradigm that combines supervised fine-tuning with multi-objective reinforcement learning to better align model behaviors with target emotions, empathetic strategies, and response formats. Extensive experiments demonstrate that STRIDE-ED generalizes across diverse open-source LLMs and consistently outperforms existing methods on both automatic metrics and human evaluations.
UR - https://www.scopus.com/pages/publications/105053633610
U2 - 10.18653/v1/2026.acl-long.321
DO - 10.18653/v1/2026.acl-long.321
M3 - 会议稿件
AN - SCOPUS:105053633610
T3 - Proceedings of the Annual Meeting of the Association for Computational Linguistics
SP - 7088
EP - 7102
BT - Long Papers
A2 - Liakata, Maria
A2 - Moreira, Viviane P.
A2 - Zhang, Jiajun
A2 - Jurgens, David
PB - Association for Computational Linguistics (ACL)
Y2 - 2 July 2026 through 7 July 2026
ER -