EmoStage: A Framework for Accurate Empathetic Response Generation via Perspective-Taking and Phase Recognition

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Qi, Zhiyang, Takamizo, Keiko, Ukiyo, Mariko, Inaba, Michimasa
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915357478879232
author Qi, Zhiyang
Takamizo, Keiko
Ukiyo, Mariko
Inaba, Michimasa
author_facet Qi, Zhiyang
Takamizo, Keiko
Ukiyo, Mariko
Inaba, Michimasa
contents The rising demand for mental health care has fueled interest in AI-driven counseling systems. While large language models (LLMs) offer significant potential, current approaches face challenges, including limited understanding of clients' psychological states and counseling stages, reliance on high-quality training data, and privacy concerns associated with commercial deployment. To address these issues, we propose EmoStage, a framework that enhances empathetic response generation by leveraging the inference capabilities of open-source LLMs without additional training data. Our framework introduces perspective-taking to infer clients' psychological states and support needs, enabling the generation of emotionally resonant responses. In addition, phase recognition is incorporated to ensure alignment with the counseling process and to prevent contextually inappropriate or inopportune responses. Experiments conducted in both Japanese and Chinese counseling settings demonstrate that EmoStage improves the quality of responses generated by base models and performs competitively with data-driven methods.
format Preprint
id arxiv_https___arxiv_org_abs_2506_19279
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle EmoStage: A Framework for Accurate Empathetic Response Generation via Perspective-Taking and Phase Recognition
Qi, Zhiyang
Takamizo, Keiko
Ukiyo, Mariko
Inaba, Michimasa
Computation and Language
Artificial Intelligence
The rising demand for mental health care has fueled interest in AI-driven counseling systems. While large language models (LLMs) offer significant potential, current approaches face challenges, including limited understanding of clients' psychological states and counseling stages, reliance on high-quality training data, and privacy concerns associated with commercial deployment. To address these issues, we propose EmoStage, a framework that enhances empathetic response generation by leveraging the inference capabilities of open-source LLMs without additional training data. Our framework introduces perspective-taking to infer clients' psychological states and support needs, enabling the generation of emotionally resonant responses. In addition, phase recognition is incorporated to ensure alignment with the counseling process and to prevent contextually inappropriate or inopportune responses. Experiments conducted in both Japanese and Chinese counseling settings demonstrate that EmoStage improves the quality of responses generated by base models and performs competitively with data-driven methods.
title EmoStage: A Framework for Accurate Empathetic Response Generation via Perspective-Taking and Phase Recognition
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2506.19279