Our research is organized into four connected areas, each with its own methodological strengths and application focus.
Computer Vision
Research on perception, visual representation learning, scene understanding, and generative vision models.
Spatial Intelligence
Research on spatial grounding, and multimodal interaction in 3D for building interactive and consistent world models.
Multi-Agent Systems
Research on cooperation, planning, communication, and decentralized intelligence across agents.
【研心】教育大模型
【研心】认知感知教育大模型研究。 Research on tuning foundation models for applications in educational and psychological cognition.
