mlsanityData to 14 Oct 2026

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

arXiv 2509.21849 ↗First posted 26 Sep 2025cs.CL cs.MAcited by 0

On the map: LLM agents and tool use → Social simulation and multi-agent debate

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use. Builds on 22 of its 29 references that are in the map. First posted on arXiv 26 Sep 2025.

Builds on

22 of 29 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025CAF-I: A Collaborative Multi-Agent Framework for Enhanced Irony Detection with Large Language Models10 Jun

Other references · 21

  1. 2023GPT-4 Technical Reportcited by 18,396
  2. 2023A Survey of Large Language Modelscited by 2,576
  3. 2015A Diversity-Promoting Objective Function for Neural Conversation Modelscited by 1,077
  4. 2022Large Language Models Are Human-Level Prompt Engineerscited by 1,045
  5. 2019DialoGPT: Large-Scale Generative Pre-training for Conversational Response Generationcited by 842
  6. 2020Recipes for building an open-domain chatbotcited by 593
  7. 2016How NOT To Evaluate Your Dialogue System: An Empirical Study of Unsupervised Evaluation Metrics for Dialogue Response Generationcited by 356
  8. 2015Building End-To-End Dialogue Systems Using Generative Hierarchical Neural Network Modelscited by 207
  9. 2020MIME: MIMicking Emotions for Empathetic Response Generationcited by 114
  10. 2019MoEL: Mixture of Empathetic Listenerscited by 91
  11. 2021CEM: Commonsense-aware Empathetic Response Generationcited by 89
  12. 2020Knowledge Bridging for Empathetic Dialogue Generationcited by 50
9 more
  1. 2023DialogueLLM: Context and Emotion Knowledge-Tuned Large Language Models for Emotion Recognition in Conversationscited by 30
  2. 2022Rethinking and Refining the Distinct Metriccited by 27
  3. 2022CASE: Aligning Coarse-to-Fine Cognition and Affection for Empathetic Response Generationcited by 18
  4. 2021Exemplars-guided Empathetic Response Generation Controlled by the Elements of Human Communicationcited by 9
  5. 2021Fine-Grained Emotion Prediction by Modeling Emotion Definitionscited by 9
  6. 2023DiffusEmp: A Diffusion Model-Based Framework with Multi-Grained Control for Empathetic Response Generationcited by 9
  7. 2023Reasoning before Responding: Integrating Commonsense-based Causality Explanation for Empathetic Response Generationcited by 9
  8. 2022CARE: Causality Reasoning for Empathetic Responses by Conditional Graph Generationcited by 7
  9. 2022Wish I Can Feel What You Feel: A Neural Approach for Empathetic Response Generationcited by 5

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024ReflectDiffu:Reflect between Emotion-intent Contagion and Mimicry for Empathetic Response Generation via a RL-Diffusion Framework
  2. 2024Cause-Aware Empathetic Response Generation via Chain-of-Thought Fine-Tuning
  3. 2025From Measurement to Expertise: Empathetic Expert Adapters for Context-Based Empathy in Conversational AI Agents
  4. 2026PRIDE: Privileged Information-enhanced Distillation for Empathetic Dialogue Generation
  5. 2025PEER: Unified Process-Outcome Reinforcement Learning for Structured Empathetic Reasoning
  6. 2026PRISM: Prosody-Integrated Multi-Agent Reasoning Framework for Empathetic Spoken Dialogue
  7. 2024EmPO: Emotion Grounding for Empathetic Response Generation through Preference Optimization
  8. 2025Sentient Agent as a Judge: Evaluating Higher-Order Social Cognition in Large Language Models