Papers

Selected publications and preprints.

Generative Control as Optimization: Time Unconditional Flow Matching for Adaptive and Robust Robotic Control

Zunzhe Zhang*, Runhan Huang*, Yicheng Liu, Shaoting Zhu, Linzhan Mou, Hang Zhao * Equal contribution

ICLR Workshops 2026 Vision-Language-Action Models · Robotic Control · Flow Matching · Imitation Learning · Out-of-Distribution Detection

GeCO turns flow-matching robotic control into adaptive optimization, improving efficiency, robustness, and out-of-distribution awareness for VLA policies.

Virtual Community: An Open World for Humans, Robots, and Society

Qinhong Zhou , Hongxin Zhang , Xiangye Lin , Zheyuan Zhang , Yutian Chen , Wenjun Liu , Zunzhe Zhang , et al.

ICLR 2026 2026 Embodied AI · Multi-Agent Simulation · Open-World Environments · Human-Robot Interaction · ICLR

An ICLR 2026 open-world platform and benchmark suite for studying human-robot coexistence, embodied social intelligence, and multi-agent planning.

Ella: Embodied Social Agents with Lifelong Memory

Hongxin Zhang, Zheyuan Zhang, Zeyuan Wang, Zunzhe Zhang, Lixing Fang, Qinhong Zhou, Chuang Gan

ICML MAS 2025 Embodied AI · LLM Agents · Lifelong Memory · Multi-Agent Systems · Foundation Models

An embodied social agent with structured lifelong memory for open-world social interaction and multi-agent decision-making.