GenRL: Multimodal-foundation world models for generalization in embodied agents

Open in new window