Zero-shot cross-modal transfer of Reinforcement Learning policies through a Global Workspace

OffRL
Re-assign community