Beyond Domain-Specific World Models: JEPA-Anything Uses 1 Recipe for 7 Fields

MarkTechPost · 1 week ago · 16

Researchers released JEPA-Anything, a domain-agnostic JEPA-based framework that uses one shared learning recipe across multiple world-modeling fields. It tested Orthogonal Predictive Factorization (OPF) with K = 4 factors per JEPA latent target (d = K × r). The approach splits the target into orthogonal subspaces, recombines factor predictions with a pseudoinverse, and yields improved matched dynamics results on all 10 tasks (including a 34.83% MSE drop on Interventional Pong single-intervention).