arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2508.04423cs.CL
收藏

面向客户支持对话的评估、合成与增强

Evaluating, Synthesizing, and Enhancing for Customer Support Conversation

Jie Zhu, Huaixia Dou, Junhui Li, Lifan Guo, Feng Chen, Chi Zhang, Fang Kong

AI总结:

本文提出客户支持对话(CSC)任务,构建基于COPC指南的五阶段十二策略框架,并利用LLM构建评估数据集CSConv和训练数据集RoleCS,微调后显著提升LLM生成策略对齐的高质量客服回复能力。

摘要:

有效的客户支持不仅需要准确解决问题,还需要符合专业标准的结构化且富有同理心的沟通。然而,现有对话数据集往往缺乏策略指导,而真实服务数据难以获取和标注。为解决这一问题,我们提出了客户支持对话(CSC)任务,旨在训练客服人员使用明确定义的支持策略进行回复。我们提出了一个基于COPC指南的结构化CSC框架,定义了五个对话阶段和十二种策略,以指导高质量交互。在此基础上,我们构建了CSConv,这是一个包含1,855个真实客户-客服对话的评估数据集,使用LLM重写以反映有意的策略使用,并进行了相应标注。此外,我们开发了一种角色扮演方法,利用与CSC框架对齐的LLM驱动角色模拟富含策略的对话,从而生成训练数据集RoleCS。实验表明,在RoleCS上微调强大的LLM可显著提高其在CSConv上生成高质量、策略对齐回复的能力。人工评估进一步证实了在问题解决方面的提升。所有代码和数据将在https://github.com/aliyun/qwen-dianjin上公开。

英文摘要:

Effective customer support requires not only accurate problem solving but also structured and empathetic communication aligned with professional standards. However, existing dialogue datasets often lack strategic guidance, and real-world service data is difficult to access and annotate. To address this, we introduce the task of Customer Support Conversation (CSC), aimed at training customer service agents to respond using well-defined support strategies. We propose a structured CSC framework grounded in COPC guidelines, defining five conversational stages and twelve strategies to guide high-quality interactions. Based on this, we construct CSConv, an evaluation dataset of 1,855 real-world customer-agent conversations rewritten using LLMs to reflect deliberate strategy use, and annotated accordingly. Additionally, we develop a role-playing approach that simulates strategy-rich conversations using LLM-powered roles aligned with the CSC framework, resulting in the training dataset RoleCS. Experiments show that fine-tuning strong LLMs on RoleCS significantly improves their ability to generate high-quality, strategy-aligned responses on CSConv. Human evaluations further confirm gains in problem resolution. All code and data will be made publicly available at https://github.com/aliyun/qwen-dianjin.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑