Mixture-of-World Models: Scaling Multi-Task Reinforcement Learning with Modular Latent Dynamics

DRLRe-assign community