Awesome-LLM-Empathy
Papers list of empathy in LMs: theory, modeling, systems, emotion, evaluation.
Last updated Jun 16, 2026
80
Stars
3
Forks
0
Issues
0
Stars/day
Attention Score
50
Topics
Language breakdown
No language data available.
โธ Files
click to expand
README
Awesome-LLM-Empathy Papers
๐งญ Exploring empathetic AI? We hope this collection proves useful in your journey. If you'd like to support the project, feel free to โญ๏ธ the repo and share it with your peers. Contributions are warmly welcome!
๐ฅ News
๐ข This list is actively maintained, and community contributions are always appreciated!
Feel free to open a pull request if you find any relevant papers.
- ๐
2026-01: Excited to share that our work, Kardia-R1_, has been accepted at WWW 2026! - ๐
2025-12: Our paper Kardia-R1_ was released on arxiv, check it now!!
- ๐
2025-05: Our paper ReflectDiffu_ was accepted at ACL 2025 Main!
๐ผ๏ธ Click to view our ReflectDiffu Poster (ACL 2025)
- ๐
2025-05: Repository launched to curate a comprehensive list of Empathy-focused LM research.
- ๐ก Theory & Frameworks - ๐ฃ๏ธ Empathetic Dialogue Systems - ๐งช Generative Approaches for Empathetic Responses - ๐ Knowledge-Enhanced Empathetic Dialogue - ๐๏ธ Controllable & Personalized Empathy - ๐ง Emotion & Empathy Understanding - ๐ญ Emotion Recognition in Conversations (ERC) - ๐งฉ Cognitive & Affective Empathy Modeling - ๐ Evaluation & Benchmarking - ๐ Applications & Use Cases - โ๏ธ Ethics & Societal Impact - ๐ Datasets - ๐ Surveys & Overviews
๐ Introduction
This repository offers a comprehensive and up-to-date collection of research papers on Empathy in Large Language Models (LLMs).As LLMs are increasingly integrated into real-world applications, their ability to express, understand, and simulate empathy is critical for building trustworthy and human-centric AI.
This list spans across:
- Theoretical frameworks
- Dialogue systems
- Cognitive modeling
- Evaluation strategies
- Applications and ethical concerns
๐ Papers
โ ๏ธ Note: To keep this README clean, the full paper list is shown below. You can also explore the papers section directly or contribute using the guidelines at the end.
๐ก Theory & Frameworks
- Affective Computing in the Era of Large Language Models: A Survey from the NLP Perspective.
- Empathy and the Right to Be an Exception: What LLMs Can and Cannot Do.
- The Illusion of Empathy: How AI Chatbots Shape Conversation Perception.
- Language Models Predict Empathy Gaps Between Social In-groups and Out-groups.
- Empathy Through Multimodality in Conversational Interfaces.
- SoulChat: Improving LLMs' Empathy, Listening, and Comfort Abilities.
- CoMAE: A Multi-factor Hierarchical Framework for Empathetic Response Generation.
๐ฃ๏ธ Empathetic Dialogue Systems
๐งช Generative Approaches for Empathetic Responses
- Kardia-R1: Unleashing LLMs to Reason toward Understanding and Empathy for Emotional Support via Rubric-as-Judge Reinforcement Learning
- ReflectDiffu: Reflect between Emotion-intent Contagion and Mimicry for Empathetic Response Generation via a RL-Diffusion Framework.
- Non-Emotion-Centric Empathetic Dialogue Generation.
- ECC: Synergizing Emotion, Cause and Commonsense for Empathetic Dialogue Generation.
- Cause-Aware Empathetic Response Generation via Chain-of-Thought Fine-Tuning.
- EmPO: Emotion Grounding for Empathetic Response Generation through Preference Optimization.
- Harnessing the Power of Large Language Models for Empathetic Response Generation.
- DiffusEmp: A Diffusion Model-Based Framework with Multi-Grained Control for Empathetic Response Generation.
- InsideOut: Unifying Emotional LLMs to Foster Empathy.
- SoulChat: Improving LLMs' Empathy, Listening, and Comfort Abilities.
- MIME: Mimicking Emotions for Empathetic Response Generation.
- CoMAE: A Multi-factor Hierarchical Framework.
- EmpDG: Multi-resolution Interactive Empathetic Dialogue Generation.
๐ Knowledge-Enhanced Empathetic Dialogue
- Sibyl: Empowering Empathetic Dialogue Generation in Large Language Models via Sensible and Visionary Commonsense Inference.
- TOOL-ED: Enhancing Empathetic Response Generation with the Tool Calling Capability of LLM.
- The Emotional Spectrum of LLMs: Leveraging Empathy and Emotion-Based Markers for Mental Health Support.
- Knowledge-enhanced Mixed-initiative Dialogue System for Emotional Support Conversations.
- CEM: Commonsense-aware Empathetic Response Generation.
๐๏ธ Controllable & Personalized Empathy
- Sibyl: Empowering Empathetic Dialogue Generation in Large Language Models via Sensible and Visionary Commonsense Inference.
- EmpCRL: Controllable Empathetic Response Generation via In-Context Reinforcement Learning.
- EmpHi: Generating Empathetic Responses with Human-like Intents.
- CARE: Causality Reasoning for Empathetic Responses by Conditional Graph Generation.
- Improving Empathetic Response Generation by Recognizing Emotion Cause in Conversations.
๐ง Emotion & Empathy Understanding
๐ญ Emotion Recognition in Conversations (ERC)
- DialogueMMT: Dialogue Scenes Understanding Enhanced Multi-modal Multi-task Tuning for Emotion Recognition in Conversations.
- DialogueLLM: Context and Emotion Knowledge-Tuned Large Language Models for Emotion Recognition in Conversations.
- Chinchunmei at WASSA 2024: Boosting LLM's Prediction with Role-play Augmentation and Contrastive Reasoning Calibration.
- Beyond Silent Letters: Amplifying LLMs in Emotion Recognition with Vocal Nuances.
- LaERC-S: Improving LLM-based Emotion Recognition in Conversation with Speaker Characteristics.
๐งฉ Cognitive & Affective Empathy Modeling
- APTNESS: Incorporating Appraisal Theory and Emotion Support Strategies for Empathetic Response Generation.
- CAB: Empathetic Dialogue Generation with Cognition, Affection and Behavior.
- MoEL: Mixture of Empathetic Listeners.
- Rational Sensibility: LLM Enhanced Empathetic Response Generation Guided by Self-presentation Theory.
- Chain of empathy: Enhancing empathetic response of large language models based on psychotherapy models.
๐ Evaluation & Benchmarking
- Apathetic or Empathetic? Evaluating LLMs' Emotional Alignments with Humans.
- EmotionQueen: A Benchmark for Evaluating Empathy of Large Language Models.
- CuLEmo: Cultural Lenses on Emotion โ Benchmarking LLMs for Cross-Cultural Emotion Understanding.
๐ Applications & Use Cases
- Customizing Emotional Support: How Do Individuals Construct and Interact With LLM-Powered Chatbots.
- From Personas to Talks: Revisiting the Impact of Personas on LLM-Synthesized Emotional Support Conversations.
- From Traits to Empathy: Personality-Aware Multimodal Empathetic Response Generation.
โ๏ธ Ethics & Societal Impact
- The Illusion of Empathy: How AI Chatbots Shape Conversation Perception.
- Language Models Predict Empathy Gaps Between Social In-groups and Out-groups.
- Empathy and the Right to Be an Exception: What LLMs Can and Cannot Do.
๐ Resources
๐ Datasets
- Towards Empathetic Open-domain Conversation Models: a New Benchmark and Dataset.
- PsyQA: A Chinese Dataset for Generating Long Counseling Text for Mental Health Support.
- TOOL-ED: Enhancing Empathetic Response Generation with the Tool Calling Capability of LLM .
- EmpathicStories++: A Multimodal Dataset for Empathy towards Personal Experiences.
- STICKERCONV: Generating Multimodal Empathetic Responses from Scratch .
- Kardia-R1: Unleashing LLMs to Reason toward Understanding and Empathy for Emotional Support via Rubric-as-Judge Reinforcement Learning
๐ Surveys & Overviews
- Affective Computing in the Era of Large Language Models: A Survey from the NLP Perspective.
๐ Contributing
โญ Help us grow this repository! If you know any valuable works weโve missed, donโt hesitate to contribute โ every suggestion makes a difference!We welcome and appreciate all contributions! Hereโs how you can help:
- ๐ Add or Update a Paper
- โ๏ธ Use Consistent Formatting
- ๐ Include Abstract Link
/abs/ link format for the abstract (e.g., https://arxiv.org/abs/xxxx.xxxxx).
- ๐ก Explain Your Edit (Optional but Helpful)
โ Don't worry about getting everything perfect!
Minor mistakes are totally fine โ weโll help fix them. What matters most is your contribution. Let's highlight your awesome work together!
Contributors
Thanks to all our amazing contributors!
๐ Citation
If you're interested in empathetic dialogue, you're welcome to take a look at our recent paper โ and feel free to cite it if you find it useful:
@inproceedings{yuan2025reflectdiffu,
title = "ReflectDiffu: Reflect between Emotion-intent Contagion and Mimicry for Empathetic Response Generation via a {RL}-Diffusion Framework",
author={Yuan, Jiahao and Di, Zixiang and Cui, Zhiqing and Yang, Guisong and Naseem, Usman},
booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
pages = "25435--25449",
year = "2025"
}
@article{yuan2025kardia, title={Kardia-R1: Unleashing LLMs to Reason toward Understanding and Empathy for Emotional Support via Rubric-as-Judge Reinforcement Learning}, author={Yuan, Jiahao and Cui, Zhiqing and Wang, Hanqing and Gao, Yuansheng and Zhou, Yucheng and Naseem, Usman}, journal={arXiv preprint arXiv:2512.01282}, year={2025} }
๐ License
This project is licensed under the MIT License.
