Following the TRACE: A Structured Path to Empathetic Response Generation with Multi-Agent Models

Ziqi Liu; Ziyang Zhou; Yilin Li; Haiyang Zhang; Yangbin Chen

Following the TRACE: A Structured Path to Empathetic Response Generation with Multi-Agent Models

Ziqi Liu
, Ziyang Zhou
, Yilin Li
, Haiyang Zhang
, Yangbin Chen^*

^*Corresponding author for this work

Department of Computing

Xi'an Jiaotong-Liverpool University

Research output: Chapter in Book or Report/Conference proceeding › Conference Proceeding › peer-review

Abstract

Empathetic response generation is a crucial task for creating more human-like and supportive conversational agents. However, existing methods face a core trade-off between the analytical depth of specialized models and the generative fluency of Large Language Models (LLMs). To address this, we propose TRACE, Task-decomposed Reasoning for Affective Communication and Empathy, a novel framework that models empathy as a structured cognitive process by decomposing the task into a pipeline for analysis and synthesis. By building a comprehensive understanding before generation, TRACE unites deep analysis with expressive generation. Experimental results show that our framework significantly outperforms strong baselines in both automatic and LLM-based evaluations, confirming that our structured decomposition is a promising paradigm for creating more capable and interpretable empathetic agents. Our code is available at https://anonymous.4open.science/r/TRACE-18EF/README.md.

Original language	English
Title of host publication	The 51th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)
Publication status	Accepted/In press - 2026

Access to Document

https://arxiv.org/abs/2509.21849

Cite this

@inproceedings{d38113da5f3a4a578336c4ff795b4b4a,

title = "Following the TRACE: A Structured Path to Empathetic Response Generation with Multi-Agent Models",

abstract = "Empathetic response generation is a crucial task for creating more human-like and supportive conversational agents. However, existing methods face a core trade-off between the analytical depth of specialized models and the generative fluency of Large Language Models (LLMs). To address this, we propose TRACE, Task-decomposed Reasoning for Affective Communication and Empathy, a novel framework that models empathy as a structured cognitive process by decomposing the task into a pipeline for analysis and synthesis. By building a comprehensive understanding before generation, TRACE unites deep analysis with expressive generation. Experimental results show that our framework significantly outperforms strong baselines in both automatic and LLM-based evaluations, confirming that our structured decomposition is a promising paradigm for creating more capable and interpretable empathetic agents. Our code is available at https://anonymous.4open.science/r/TRACE-18EF/README.md.",

author = "Ziqi Liu and Ziyang Zhou and Yilin Li and Haiyang Zhang and Yangbin Chen",

year = "2026",

language = "English",

booktitle = "The 51th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)",

}

TY - GEN

T1 - Following the TRACE: A Structured Path to Empathetic Response Generation with Multi-Agent Models

AU - Liu, Ziqi

AU - Zhou, Ziyang

AU - Li, Yilin

AU - Zhang, Haiyang

AU - Chen, Yangbin

PY - 2026

Y1 - 2026

N2 - Empathetic response generation is a crucial task for creating more human-like and supportive conversational agents. However, existing methods face a core trade-off between the analytical depth of specialized models and the generative fluency of Large Language Models (LLMs). To address this, we propose TRACE, Task-decomposed Reasoning for Affective Communication and Empathy, a novel framework that models empathy as a structured cognitive process by decomposing the task into a pipeline for analysis and synthesis. By building a comprehensive understanding before generation, TRACE unites deep analysis with expressive generation. Experimental results show that our framework significantly outperforms strong baselines in both automatic and LLM-based evaluations, confirming that our structured decomposition is a promising paradigm for creating more capable and interpretable empathetic agents. Our code is available at https://anonymous.4open.science/r/TRACE-18EF/README.md.

AB - Empathetic response generation is a crucial task for creating more human-like and supportive conversational agents. However, existing methods face a core trade-off between the analytical depth of specialized models and the generative fluency of Large Language Models (LLMs). To address this, we propose TRACE, Task-decomposed Reasoning for Affective Communication and Empathy, a novel framework that models empathy as a structured cognitive process by decomposing the task into a pipeline for analysis and synthesis. By building a comprehensive understanding before generation, TRACE unites deep analysis with expressive generation. Experimental results show that our framework significantly outperforms strong baselines in both automatic and LLM-based evaluations, confirming that our structured decomposition is a promising paradigm for creating more capable and interpretable empathetic agents. Our code is available at https://anonymous.4open.science/r/TRACE-18EF/README.md.

M3 - Conference Proceeding

BT - The 51th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

ER -