Acknowledgment of Emotional States: Generating Validating Responses for Empathetic Dialogue

TL;DR

This paper presents a novel framework for generating validating responses in empathetic human-AI dialogue, utilizing a tripartite module system and outperforming baseline models on multiple datasets.

Contribution

Introduces the first framework for empathetic dialogue with validating responses, combining validation timing detection, emotional state identification, and response generation.

Findings

Model outperforms baselines on Japanese EmpatheticDialogues dataset.

Model surpasses ChatGPT in F1-score across modules.

Framework effective in both textual and speech-based dialogues.

Abstract

In the realm of human-AI dialogue, the facilitation of empathetic responses is important. Validation is one of the key communication techniques in psychology, which entails recognizing, understanding, and acknowledging others' emotional states, thoughts, and actions. This study introduces the first framework designed to engender empathetic dialogue with validating responses. Our approach incorporates a tripartite module system: 1) validation timing detection, 2) users' emotional state identification, and 3) validating response generation. Utilizing Japanese EmpatheticDialogues dataset - a textual-based dialogue dataset consisting of 8 emotional categories from Plutchik's wheel of emotions - the Task Adaptive Pre-Training (TAPT) BERT-based model outperforms both random baseline and the ChatGPT performance, in term of F1-score, in all modules. Further validation of our model's efficacy is confirmed in its application to the TUT Emotional Storytelling Corpus (TESC), a speech-based dialogue dataset, by surpassing both random baseline and the ChatGPT. This consistent performance across both textual and speech-based dialogues underscores the effectiveness of our framework in fostering empathetic human-AI communication.