README.md

metadata

{}

🤖 Neural Motion Simulator

Official release of the model checkpoints from the paper:
Neural Motion Simulator: Pushing the Limit of World Models in Reinforcement Learning

The released model consists of two parts: the main world model (MoSim model) and a flow model used to implement the functionality described in Section 3.7 of the paper. All MoSim models are trained exclusively on data collected from random policies.