AI RESEARCH
The Global Neural World Model: Spatially Grounded Discrete Topologies for Action-Conditioned Planning
arXiv CS.LG
•
ArXi:2604.16585v1 Announce Type: new We present the Global Neural World Model (GNWM), a self-stabilizing framework that achieves topological quantization through balanced continuous entropy constraints. Operating as a continuous, action-conditioned Joint-Embedding Predictive Architecture (JEPA), the GNWM maps environments onto a discrete 2D grid, enforcing translational equivariance without pixel-level reconstruction. Our results show this architecture prevents manifold drift during autoregressive rollouts by using grid ``snapping'' as a native error-correction mechanism. Furthermore, by