Our paper develops training-free world models based on foundation models with interpretable latent states.
Update: presented at the probabilistic robotics workshop at ICRA’24.
Citation:
- Zhenjiang Mao, Siqi Dai, Yuang Geng, Ivan Ruchkin.
Zero-shot Safety Prediction for Autonomous Robots with Foundation World Models [Arxiv] [Poster].
Back to the Future: Robot Learning Going Probabilistic Workshop (co-located with ICRA 2024), Yokohama, Japan, 2024.