embodied intelligence
Embodied intelligence refers to the concept of intelligence being leveraged through a physical form or body, allowing agents to interact with their environment and learn from those interactions. This concept is foundational for robotics and AI systems that operate in the real world.
- DiffE2E: Rethinking End-to-End Driving with a Hybrid Diffusion-Regression-Classification Policy
- EfficientVLA: Training-Free Acceleration and Compression for Vision-Language-Action Models
- Enhancing LLM Planning for Robotics Manipulation through Hierarchical Procedural Knowledge Graphs
- From Pose to Muscle: Multimodal Learning for Piano Hand Muscle Electromyography
- Intend to Move: A Multimodal Dataset for Intention-Aware Human Motion Understanding
- LabUtopia: High-Fidelity Simulation and Hierarchical Benchmark for Scientific Embodied Agents
- MV-CoLight: Efficient Object Compositing with Consistent Lighting and Shadow Generation
- OpenMMEgo: Enhancing Egocentric Understanding for LMMs with Open Weights and Data
- RoboScape: Physics-informed Embodied World Model
- SEEA-R1: Tree-Structured Reinforcement Fine-Tuning for Self-Evolving Embodied Agents
- VolleyBots: A Testbed for Multi-Drone Volleyball Game Combining Motion Control and Strategic Play