EmPO：通过偏好优化实现共情响应生成的情绪锚定
EmPO: Emotion Grounding for Empathetic Response Generation through Preference Optimization
AI总结：
本研究提出EmPO方法，基于情绪锚定构建理论驱动的偏好数据集，结合偏好优化算法对齐大语言模型，在提升共情响应生成质量的同时保留模型泛化性能，并公开相关资源。
摘要：
共情响应生成是对话智能体的一项理想特性，对促进人机之间引人入胜且具备情绪智能的多轮对话至关重要。利用大语言模型完成该任务已展现出可观成效，但在保障响应的共情质量与保留模型泛化性能两方面仍存在挑战。我们提出一种新方法：基于情绪锚定构建理论驱动的偏好数据集，借助偏好优化算法对齐大语言模型，以此解决上述挑战。为评估共情响应生成效果，我们采用EmpatheticDialogues数据集，通过diff-Epitome、BERTscore指标及多维度人工评估衡量共情水平，另外使用基于特征的方法测量多样性和情绪效价。我们还通过MMLU基准以及Open LLM Leaderboard中的任务评估训练对泛化性能的影响。结果表明，大语言模型可通过偏好优化实现共情响应生成的对齐，同时保留其通用性能，且情绪锚定能够指导偏好数据集的构建。我们公开了所有数据集、源代码和模型，详见https://github.com/justtherightsize/empo。
英文摘要：
Empathetic response generation is a desirable aspect of conversational agents, crucial for facilitating engaging and emotionally intelligent multi-turn conversations between humans and machines. Leveraging large language models for this task has shown promising results, yet challenges persist in ensuring both the empathetic quality of the responses and retention of the generalization performance of the models. We propose a novel approach where we construct theory-driven preference datasets based on emotion grounding and use them to align LLMs with preference optimization algorithms to address these challenges. To evaluate empathetic response generation, we employ the EmpatheticDialogues dataset, assessing empathy with the diff-Epitome and BERTscore metrics and with multi-dimensional human evaluation. Additionally, we measure diversity and emotional valence using feature-based methods. We also evaluate the impact of training on the generalization performance using the MMLU benchmark and tasks from the Open LLM Leaderboard. The results show that LLMs can be aligned for empathetic response generation by preference optimization while retaining their general performance and that emotion grounding can guide preference dataset creation. We make all datasets, source code, and models publicly available. https://github.com/justtherightsize/empo
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。
