A new benchmark called RoboDojo has been released to evaluate embodied AI, featuring 42 simulation tasks and 18 real-world robot tasks. The benchmark highlights a significant gap between current AI models and human performance, with the best models achieving only 8.80% success rate in simulation and 12.8% on real robots, compared to humans' 76.03% and 100% respectively. RoboDojo aims to provide a standardized and comprehensive evaluation for embodied intelligence, covering generalization, memory, precision, long-horizon execution, and open-ended semantic understanding. AI
IMPACT Highlights the significant gap in current AI capabilities for real-world robotics, indicating a need for more robust models.
RANK_REASON The cluster describes a new benchmark and evaluation framework for embodied AI, including research findings on model performance. [lever_c_demoted from research: ic=1 ai=1.0]
- ARX X5
- GalaxeaVLA
- GR00T-N1.7
- Hy-Embodied-0.5-VLA
- InternVLA-A1
- OpenVLA-OFT
- Piper
- RoboDojo
- RoboTwin
- Xiaomi-Robotics-0
- XPolicyLab
- X-VLA
- π0
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
