Dialogue Systems for Emotional Support via Value Reinforcement
Juhee Kim, Chunghu Mok, Jisun Lee, Hyang Sook Kim, Yohan Jo
摘要
Emotional support dialogue systems aim to reduce help-seekers' distress and help them overcome challenges. While human values-core beliefs that shape an individual's prioritiesare increasingly emphasized in contemporary psychological therapy for their role in fostering internal transformation and long-term emotional well-being, their integration into emotional support systems remains underexplored. To bridge this gap, we present a value-driven method for training emotional support dialogue systems designed to reinforce positive values in seekers. Notably, our model identifies which values to reinforce at each turn and how to do so, by leveraging online support conversations from Reddit. We evaluate the method across support skills, seekers' emotional intensity, and value reinforcement. Our method consistently outperforms various baselines, effectively exploring and eliciting values from seekers. Additionally, leveraging crowd knowledge from Reddit significantly enhances its effectiveness. Therapists highlighted its ability to validate seekers' challenges and emphasize positive aspects of their situations-both crucial elements of value reinforcement. Our work, being the first to integrate value reinforcement into emotional support systems, demonstrates its promise and establishes a foundation for future research.
问问这篇 Paper
智能体会读完全文。
Lune 把这篇 Paper 索引到了每一个公式,引用它的顶会 Paper 也一样。你提问,回答直接引用原文。
引用它的顶会 Paper2
- Act-Adaptive Margin: Dynamically Calibrating Reward Models for Subjective AmbiguityFeiteng Fang, Dingwei Chen, Xiang Huang, Ting-En Lin 等ACL 2026 · 被引用 3 次
- Discourse Coherence and Response-Guided Context Rewriting for Multi-Party Dialogue GenerationZhiyu Cao, Peifeng Li, Qiaoming ZhuACL 2026
它引用的顶会 Paper12
- LoRA: Low-Rank Adaptation of Large Language ModelsEdward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu 等ICLR 2022 · 被引用 18,833 次
- Direct Preference Optimization: Your Language Model is Secretly a Reward ModelRafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D. Manning 等NeurIPS 2023 · 被引用 10,924 次
- Plug-and-Play Policy Planner for Large Language Model Powered Dialogue AgentsYang Deng, Wenxuan Zhang, Wai Lam, See-Kiong Ng 等ICLR 2024 · 被引用 86 次
- Improving Multi-turn Emotional Support Dialogue Generation with Lookahead Strategy PlanningYi Cheng, Wenge Liu, Wenjie Li, Jiashuo Wang 等EMNLP 2022 · 被引用 32 次
- Facilitating Multi-turn Emotional Support Conversation with Positive Emotion Elicitation: A Reinforcement Learning ApproachJinfeng Zhou, Zhuang Chen, Bo Wang, Minlie HuangACL 2023 · 被引用 17 次
相关 Paper
- SocialSim: Towards Socialized Simulation of Emotional Support ConversationZhuang Chen, Yaru Cao, Guanqun Bi, Jincenzi Wu 等AAAI 2025 · 被引用 12 次
- ESCA: An Emotional Support Conversation Agent for Enhancing Reasonable Strategy Planning and Effective ExpressionJing Li, Yanxin Luo, Donghong Han, Yimeng Zhan 等AAAI 2026
- Developing a Social Support Framework: Understanding the Reciprocity in Human-Chatbot RelationshipShuyi Pan, Maartje M. A. de GraafCHI 2025 · 被引用 19 次
- Towards Emotional Support Dialog SystemsSiyang Liu, Chujie Zheng, Orianna Demasi, Sahand Sabour 等ACL 2021
- ValueNet: A New Dataset for Human Value Driven Dialogue SystemLiang Qiu, Yizhou Zhao, Jinchao Li, Pan Lu 等AAAI 2022 · 被引用 51 次
