TY - GEN
T1 - Chain of Strategy Optimization Makes Large Language Models Better Emotional Supporter
AU - Zhao, Weixiang
AU - Sui, Xingyu
AU - Han, Xinyang
AU - Deng, Yang
AU - Hu, Yulin
AU - Guo, Jiahe
AU - Qin, Libo
AU - Du, Qianyun
AU - Wang, Shijin
AU - Zhao, Yanyan
AU - Qin, Bing
AU - Liu, Ting
N1 - Publisher Copyright:
©2025 Association for Computational Linguistics.
PY - 2025
Y1 - 2025
N2 - The growing emotional stress in modern society has increased the demand for Emotional Support Conversations (ESC). While Large Language Models (LLMs) show promise for ESC, they face two key challenges: (1) low strategy selection accuracy, and (2) preference bias, limiting their adaptability to users’ emotional needs. Existing supervised fine-tuning (SFT) struggles to address these issues, as it rigidly trains models on single gold-standard responses without modeling nuanced strategy trade-offs. To overcome these limitations, we propose a novel two-stage framework that optimizes strategy selection preferences at each dialogue turn. We first leverage Monte Carlo Tree Search to construct ESC-Pro, a high-quality preference dataset with turn-level strategy-response pairs. Then training on ESC-Pro with Chain-of-Strategy Optimization (CSO) improves both strategy accuracy and bias mitigation, enabling LLMs to generate more empathetic and contextually appropriate responses. Experiments on LLaMA-3.1-8B, Gemma-2-9B, and Qwen2.5-7B demonstrate that CSO outperforms standard SFT, highlighting the efficacy of fine-grained, turn-level preference modeling in ESC.1
AB - The growing emotional stress in modern society has increased the demand for Emotional Support Conversations (ESC). While Large Language Models (LLMs) show promise for ESC, they face two key challenges: (1) low strategy selection accuracy, and (2) preference bias, limiting their adaptability to users’ emotional needs. Existing supervised fine-tuning (SFT) struggles to address these issues, as it rigidly trains models on single gold-standard responses without modeling nuanced strategy trade-offs. To overcome these limitations, we propose a novel two-stage framework that optimizes strategy selection preferences at each dialogue turn. We first leverage Monte Carlo Tree Search to construct ESC-Pro, a high-quality preference dataset with turn-level strategy-response pairs. Then training on ESC-Pro with Chain-of-Strategy Optimization (CSO) improves both strategy accuracy and bias mitigation, enabling LLMs to generate more empathetic and contextually appropriate responses. Experiments on LLaMA-3.1-8B, Gemma-2-9B, and Qwen2.5-7B demonstrate that CSO outperforms standard SFT, highlighting the efficacy of fine-grained, turn-level preference modeling in ESC.1
UR - https://www.scopus.com/pages/publications/105028944071
U2 - 10.18653/v1/2025.findings-emnlp.831
DO - 10.18653/v1/2025.findings-emnlp.831
M3 - 会议稿件
AN - SCOPUS:105028944071
T3 - EMNLP 2025 - 2025 Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP 2025
SP - 15361
EP - 15381
BT - EMNLP 2025 - 2025 Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP 2025
A2 - Christodoulopoulos, Christos
A2 - Chakraborty, Tanmoy
A2 - Rose, Carolyn
A2 - Peng, Violet
PB - Association for Computational Linguistics (ACL)
T2 - 30th Conference on Empirical Methods in Natural Language Processing, EMNLP 2025
Y2 - 4 November 2025 through 9 November 2025
ER -