About the Journal
Large Models and Embodied Intelligence is an open-access journal dedicated to advancing frontier research in large-scale models and embodied intelligence. The journal aims to promote the paradigm shift of artificial intelligence from static perception and reasoning toward an integrated framework encompassing perception, cognition, decision-making, action, and interaction. The journal focuses on the deep integration of large foundation models, intelligent agents, and physical-world interaction, with an emphasis on autonomy, generalization, and safety in real-world or high-fidelity environments. It covers theoretical foundations, model architectures, learning paradigms, system implementations, and practical applications of embodied intelligence, and encourages original contributions that bridge artificial intelligence, robotics, control, cognitive science, and human–computer interaction. Large Models and Embodied Intelligence welcomes high-quality original research article, review, perspective, commentary and more types. The journal emphasizes both methodological innovation and validation in real-world settings, addressing key challenges toward general-purpose embodied intelligence.
Topics of interest include, but are not limited to:
- Large-scale pre-trained models (LLMs, VLMs, VLAs)
- Embodied agents and multi-agent systems
- Multimodal embodied perception and spatiotemporal understanding
- Embodied manipulation and control
- Embodied navigation and motion planning
- Embodied world models and causal reasoning
- Reinforcement learning and imitation learning
- Embodied autonomous and lifelong learning
- Human–robot interaction and collaboration
- Collective and swarm embodied intelligence
- Applications of Large Models
- Safety, trustworthiness, and ethics of embodied intelligence
