I am a second-year Ph.D. student at Peking University. My research focuses on socially intelligent agents: how they understand social interactions, make decisions in multi-agent environments, and interact effectively with humans and other agents. I want to build learning algorithms, benchmarks, and diagnostic methods with the goal of making agent systems more adaptive, reliable, and socially aware.
Research interests: multi-agent systems, social intelligence, large language model agents, agent evaluation and learning
Email: kfq20@stu.pku.edu.cn · Phone: (+86) 173-5819-0815 · Location: Beijing, China
🔥 News
- 2026: 🎉 Three papers were accepted: SIV-Bench at Findings of ACL 2026, InfoPO at the Lifelong Agent Workshop @ ICLR 2026, and Aegis at ICLR 2026.
📝 Publications
SIV-Bench: A Video Benchmark for Social Interaction Understanding and Reasoning
Fanqi Kong, Weiqin Zu, Xinyu Chen, Yaodong Yang, Song-Chun Zhu, Xue Feng
Paper · Code · Dataset
- A video benchmark for evaluating social scene understanding, social state reasoning, and social dynamics prediction in multimodal models.
InfoPO: Information-Driven Policy Optimization for User-Centric Agents
Fanqi Kong, Jiayi Zhang, Mingyi Deng, Chenglin Wu, Yuyu Luo, Bang Liu
Paper · Code
- Optimizes user-centric agents by rewarding information-seeking interactions that reduce uncertainty and improve downstream decisions.
Aegis: Automated Error Generation and Attribution for Multi-Agent Systems
Fanqi Kong, Ruijie Zhang, Huaxiao Yin, Guibin Zhang, Xiaofei Zhang, Ziang Chen, Zhaowei Zhang, Xiaoyuan Zhang, Song-Chun Zhu, Xue Feng
Paper · Code · Dataset
- An automated framework for generating diverse multi-agent failures and training models to attribute errors to the responsible agents.
Enhancing LLM-Based Social Bot via an Adversarial Learning Framework
Fanqi Kong, Xiaoyuan Zhang, Xinyu Chen, Yaodong Yang, Song-Chun Zhu, Xue Feng
Paper · Code
Oral Presentation
- Adversarial training pipeline that strengthens conversational robustness and social reasoning in LLM agents.
Learning to Balance Altruism and Self-interest Based on Empathy in Mixed-Motive Games
Fanqi Kong, Yizhe Huang, Song-Chun Zhu, Siyuan Qi, Xue Feng
Paper
- Empathetic policy learning framework that adapts cooperation dynamics in mixed-motive games.
AdaSociety: An Adaptive Environment with Social Structures for Multi-Agent Decision-Making
Yizhe Huang, Xingbo Wang, Hao Liu, Fanqi Kong, Aoyang Qin, Min Tang, Song-Chun Zhu, Mingjie Bi, Siyuan Qi, Xue Feng
Paper
- Introduces a configurable benchmark with dynamic social roles for evaluating cooperative intelligence.
Efficient Adaptation in Mixed-Motive Environments via Hierarchical Opponent Modeling and Planning
Yizhe Huang, Anji Liu, Fanqi Kong, Yaodong Yang, Song-Chun Zhu, Xue Feng
Paper
- Two-tier opponent modeling that accelerates adaptation under shifting cooperative incentives.
🧪 Preprints
AutoEnv: Automated Environments for Measuring Cross-Environment Agent Learning
Jiayi Zhang, Yiran Peng, Fanqi Kong, Yang Cheng, Yifan Wu, Zhaoyang Yu, Jinyu Xiang, Jianhao Ruan, Jinlin Wang, Maojia Song, HongZhang Liu, Xiangru Tang, Bang Liu, Chenglin Wu, Yuyu Luo
Paper · Code
- An automated environment framework for measuring how well agents learn and adapt across diverse environments with different rules and dynamics.
🏆 Honors and Awards
- 2025 — National Scholarship, Peking University.
- 2024 — Lixin Scholarship, School of Intelligence Science and Technology, Peking University.
- 2024 — Outstanding Graduate, Department of Automation, Tsinghua University.
- 2023 — Friend of Tsinghua – Hande Scholarship.
- 2023 — Tsinghua University Scholarship (Excellence in Arts and Literature).
- 2023 — Person of the Year, Department of Automation, Tsinghua University.
- 2021 — Tsinghua University Scholarship (Excellence in Social Work).
🎓 Education
- 2024 – Present — Ph.D., Institute for Artificial Intelligence, Peking University.
- 2020 – 2024 — B.S., Department of Automation, Tsinghua University.
🤝 Leadership and Service
- 2026 — Reviewer, ICML 2026.
- 2026 — Reviewer, ICLR 2026.
- 2025 — Reviewer, NeurIPS 2025.
- 2024 – Present — Head of Academic Department, School of Intelligence Science and Technology, Peking University.
- 2023 – 2024 — Presidium Member, Student Union, Department of Automation, Tsinghua University.
🎮 Competitions
- 2023 — NeurIPS Melting Pot Contest, Honorable Winner (6/117). Core contributor on mixed-motive multi-agent cooperation tasks.