DianJin-CSC-Data

<div align="center">
<img alt="image" src="https://raw.githubusercontent.com/aliyun/qwen-dianjin/refs/heads/master/images/dianjin_logo.png">
<p align="center">
<a href="https://tongyi.aliyun.com/dianjin">Qwen DianJin Platform</a> |
<a href="https://github.com/aliyun/qwen-dianjin">Github</a> |
<a href="https://modelscope.cn/organization/tongyi_dianjin">ModelScope</a> |
<a href="https://arxiv.org/abs/2508.04423">Paper</a>
</p>
</div>

## 📢 Introduction<a name="summary"></a>


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 realworld 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 supporters 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

We open-source both the CSConv and RoleCS datasets to support research on customer support conversation systems. These resources are intended to facilitate model development, benchmarking, and further advances in the field.

## 🔖 Citation<a name="cite"></a>

If you use our dataset, please cite our paper.

```
@article{dianjin-csc,
title = {Evaluating, Synthesizing, and Enhancing for Customer Support Conversation},
author = {Jie Zhu, Huaixia Dou, Junhui Li, Lifan Guo, Feng Chen, Chi Zhang, and Fang Kong},
journal = {arxiv},
year = {2025}
}
```

<div align="center">
<img alt="标识" src="https://raw.githubusercontent.com/aliyun/qwen-dianjin/refs/heads/master/images/dianjin_logo.png">
<p align="center">
<a href="https://tongyi.aliyun.com/dianjin">通义点金平台</a> |
<a href="https://github.com/aliyun/qwen-dianjin">GitHub</a> |
<a href="https://modelscope.cn/organization/tongyi_dianjin">ModelScope</a> |
<a href="https://arxiv.org/abs/2508.04423">论文</a>
</p>
</div>

## 📢 简介<a name="summary"></a>


优质的客户服务不仅需要精准解决问题，还需遵循专业规范，采用结构化且共情的沟通方式。然而，现有对话数据集往往缺乏策略层面的指导，且真实服务数据难以获取与标注。为此，我们提出了客户服务对话（Customer Support Conversation, CSC）任务，旨在训练客服人员使用规范的服务策略进行应答。我们基于COPC指南构建了结构化的CSC框架，定义了五个对话阶段与十二种策略，以指导高质量的交互过程。基于此，我们构建了CSConv评测数据集，该数据集包含1855条真实的客服-客户对话，经大语言模型（Large Language Model, LLM）改写以体现刻意使用的策略，并完成了对应标注。此外，我们开发了一种角色扮演方法，通过适配CSC框架的大语言模型驱动角色，模拟富含策略的对话，由此生成训练数据集RoleCS。实验结果表明，在RoleCS上对高性能大语言模型进行微调，可显著提升其在CSConv上生成高质量且契合策略的应答的能力。人工评估进一步证实了其在问题解决方面的性能提升。

我们开源了CSConv与RoleCS两个数据集，以支撑客户服务对话系统相关研究。本数据集旨在助力模型开发、基准测试以及该领域的进一步发展。

## 🔖 引用<a name="cite"></a>

若您使用本数据集，请引用我们的论文。

@article{dianjin-csc,
title = {Evaluating, Synthesizing, and Enhancing for Customer Support Conversation},
author = {Jie Zhu, Huaixia Dou, Junhui Li, Lifan Guo, Feng Chen, Chi Zhang, and Fang Kong},
journal = {arxiv},
year = {2025}
}

搜集汇总

背景与挑战

背景概述

DianJin-CSC-Data是一个客户支持对话数据集，包含CSConv和RoleCS两部分，旨在通过结构化框架和策略指导提升客户服务质量。该数据集基于COPC指南，定义了五个对话阶段和十二种策略，适用于模型训练和评估。