World Modeling through Spectral Alignment
arXiv:2610.15429v2 Announce Type: replace-cross Abstract: How do we tell a world model what information to preserve? We study this question through the lens of spectral representation learning. We introduce SpecWM, a JEPA framework that explicitly specifies desired relationships between observations through a target similarity kernel and trains the encoder to reproduce those relationships in its embedding geometry. Different kernels provide different supervision, ranging from physical state inf…