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. In this paper, we propose a novel
approach where we construct theory-driven preference datasets and use them to
align LLMs with preference optimization algorithms to address these challenges.
To measure empathetic response generation, we employ the EmpatheticDialogues
dataset, assessing empathy with the diff-EPITOME and BERTscore metrics, and
evaluate the generalization performance on the MMLU benchmark. We make all
datasets, source code, and models publicly available.

我们提出了一种新方法，利用理论驱动的偏好数据集和优化算法来对齐 LLMs，以提高情感回应生成的质量和模型的泛化性能，并通过 EmpatheticDialogues 数据集以及 diff-EPITOME 和 BERTscore 指标来评估其效果。

情感回应生成的理论驱动数据集构建与优化

EmPO: Theory-Driven Dataset Construction for Empathetic Response  Generation through Preference Optimization

Large Language Models (LLMs) have demonstrated remarkable performance across
various information-seeking and reasoning tasks. These computational systems
drive state-of-the-art dialogue systems, such as ChatGPT and Bard. They also
carry substantial promise in meeting the growing demands of mental health care,
albeit relatively unexplored. As such, this study sought to examine LLMs'
capability to generate empathetic responses in conversations that emulate those
in a mental health counselling setting. We selected five LLMs: version 3.5 and
version 4 of the Generative Pre-training (GPT), Vicuna FastChat-T5, Pathways
Language Model (PaLM) version 2, and Falcon-7B-Instruct. Based on a simple
instructional prompt, these models responded to utterances derived from the
EmpatheticDialogues (ED) dataset. Using three empathy-related metrics, we
compared their responses to those from traditional response generation dialogue
systems, which were fine-tuned on the ED dataset, along with human-generated
responses. Notably, we discovered that responses from the LLMs were remarkably
more empathetic in most scenarios. We position our findings in light of
catapulting advancements in creating empathetic conversational systems.

本研究通过比较五款大型语言模型（LLMs）生成的相应和传统对话系统及人工生成的相应，发现 LLMs 在大多数情景中表现出更高的同理心，从而探索并验证 LLMs 在模拟心理咨询对话中生成同理心回应的能力。

利用大型语言模型的共情回应生成能力支持在线心理健康咨询

Harnessing Large Language Models' Empathetic Response Generation  Capabilities for Online Mental Health Counselling Support

Empathetic response generation aims to comprehend the user emotion and then
respond to it appropriately. Most existing works merely focus on what the
emotion is and ignore how the emotion is evoked, thus weakening the capacity of
the model to understand the emotional experience of the user for generating
empathetic responses. To tackle this problem, we consider the emotional
causality, namely, what feelings the user expresses (i.e., emotion) and why the
user has such feelings (i.e., cause). Then, we propose a novel graph-based
model with multi-hop reasoning to model the emotional causality of the
empathetic conversation. Finally, we demonstrate the effectiveness of our model
on EMPATHETICDIALOGUES in comparison with several competitive models.

该研究旨在解决现有研究仅关注用户情绪，而忽略情绪产生原因的问题，提出了一种基于图的多跳推理模型来建模共情对话的情感因果关系，并在 EMPATHETICDIALOGUES 上与几种竞争模型进行了比较实验，验证了该模型的有效性。