Guanqun Bi

Here is Guanqun Bi (毕冠群) 👋.

I am a Postdoctoral Researcher at CoAI Group, Tsinghua University, working with Prof. Minlie Huang.

I obtained my Ph.D. degree from Institute of Information Engineering, Chinese Academy of Sciences in June 2025, where I was fortunate to be advised by Prof. Yanan Cao.

When I’m not doing research, you can find me birdwatching 🐦 in parks around Beijing!

Research Interests

My research focuses on building AI systems that truly understand and support humans:

🧠 Social Intelligence - Developing AI with human-like social understanding and interaction capabilities
💚 AI for Mental Health - Leveraging AI to support mental well-being and provide accessible care
🗣️ Dialogue Systems - Building AI that can have meaningful, empathetic conversations
✍️ Controllable Text Generation - Generating text with desired attributes and safety guarantees

My long-term goal is to build socially intelligent AI systems that can genuinely understand human emotions, provide meaningful support, and contribute to mental health and well-being.

Education & Experience

🔬 Jul. 2025 - Present: Postdoctoral Researcher CoAI Group, Tsinghua University, Beijing, China Advisor: Prof. Minlie Huang
🎓 Sept. 2019 - Jun. 2025: Ph.D. in Cyber Security (NLP direction) Institute of Information Engineering, CAS, Beijing, China Advisor: Prof. Yanan Cao
🎓 Sept. 2015 - Jun. 2019: B.E. in Software Engineering
Shandong University, Jinan, China

news

Nov 08, 2025	🎉 Our paper “Unveiling the Landscape of Clinical Depression Assessment” is accepted to AAAI 2026!
Aug 20, 2025	📝 Our paper “A Group Fairness Lens for Large Language Models” is accepted to EMNLP 2025 Findings!
May 15, 2025	📝 Our paper “MAGI: Multi-Agent Guided Interview for Psychiatric Assessment” is accepted to ACL 2025 Findings!
Dec 09, 2024	🎉 Three papers accepted to AAAI 2025: CharacterBench, SocialSim, and SS-GEN!
Jul 10, 2024	🎓 Started my postdoc position at CoAI Group, Tsinghua University, working with Prof. Minlie Huang!

selected publications

AAAI

Unveiling the Landscape of Clinical Depression Assessment: From Behavioral Signatures to Psychiatric Reasoning

Zhuang Chen, Guanqun Bi, Wen Zhang, and 5 more authors

In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2026

Abs arXiv Bib

@inproceedings{chen2026unveiling,
  title = {Unveiling the Landscape of Clinical Depression Assessment: From Behavioral Signatures to Psychiatric Reasoning},
  author = {Chen, Zhuang and Bi, Guanqun and Zhang, Wen and Hu, Jiawei and Wang, Aoyun and Xiao, Xiyao and Feng, Kun and Huang, Minlie},
  booktitle = {Proceedings of the AAAI Conference on Artificial Intelligence (AAAI)},
  year = {2026},
}

ACL

MAGI: Multi-Agent Guided Interview for Psychiatric Assessment

Guanqun Bi, Zhuang Chen, Zhoufu Liu, and 8 more authors

In Findings of the Association for Computational Linguistics: ACL 2025, 2025

Abs arXiv Bib

@inproceedings{bi2025magi,
  title = {MAGI: Multi-Agent Guided Interview for Psychiatric Assessment},
  author = {Bi, Guanqun and Chen, Zhuang and Liu, Zhoufu and Wang, Hongkai and Xiao, Xiyao and Xie, Yuqiang and Zhang, Wen and Huang, Yongkang and Chen, Yuxuan and Peng, Libiao and others},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2025},
  pages = {24898--24921},
  year = {2025},
}

EMNLP

A Group Fairness Lens for Large Language Models

Guanqun Bi, Lei Shen, Yuqiang Xie, and 3 more authors

In Findings of the Association for Computational Linguistics: EMNLP 2025, 2025

Abs arXiv Bib

@inproceedings{bi2025fairness,
  title = {A Group Fairness Lens for Large Language Models},
  author = {Bi, Guanqun and Shen, Lei and Xie, Yuqiang and Cao, Yanan and Zhu, Tiangang and He, Xiaodong},
  booktitle = {Findings of the Association for Computational Linguistics: EMNLP 2025},
  year = {2025},
}

ACL

ToMBench: Benchmarking Theory of Mind in Large Language Models

Zhuang Chen, Jincenzi Wu, Jinfeng Zhou, and 8 more authors

In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL), 2024

Abs arXiv Bib

@inproceedings{chen2024tombench,
  title = {ToMBench: Benchmarking Theory of Mind in Large Language Models},
  author = {Chen, Zhuang and Wu, Jincenzi and Zhou, Jinfeng and Wen, Bosi and Bi, Guanqun and Jiang, Gongyao and Cao, Yaru and Hu, Mengting and Lai, Yunghwei and Xiong, Zexuan and others},
  booktitle = {Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (ACL)},
  pages = {15959--15983},
  year = {2024},
}

ACL
DiffusEmp: A Diffusion Model-Based Framework with Multi-Grained Control for Empathetic Response Generation

Guanqun Bi, Lei Shen, Yanan Cao, and 3 more authors

In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL), 2023

Abs arXiv Bib

We design a framework DiffusEmp based on conditional diffusion language model to unify the utilization of dialogue context and attribute-oriented control signals for empathetic response generation.
@inproceedings{bi2023diffusemp, title = {DiffusEmp: A Diffusion Model-Based Framework with Multi-Grained Control for Empathetic Response Generation}, author = {Bi, Guanqun and Shen, Lei and Cao, Yanan and Chen, Meng and Xie, Yuqiang and He, Xiaodong}, booktitle = {Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (ACL)}, year = {2023}, }
ICASSP
SERI: SkEtching-Reasoning-Integrating Progressive Workflow for Empathetic Response Generation

Guanqun Bi, Yanan Cao, Piji Li, and 3 more authors

In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2023

Abs Bib

We propose a SkEtching-Reasoning-Integrating framework for empathetic response generation with multi-source information reasoning.
@inproceedings{bi2023seri, title = {SERI: SkEtching-Reasoning-Integrating Progressive Workflow for Empathetic Response Generation}, author = {Bi, Guanqun and Cao, Yanan and Li, Piji and Xie, Yuqiang and Fang, Fang and Lin, Zheng}, booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, year = {2023}, }