BeliefGraph-JEPA: Structured Latent World Models for Action-Conditioned Time Series

Excerpt

arXiv:2610.05409v1 Announce Type: cross Abstract: Action-conditioned time-series forecasting requires accounting for how future actions and exogenous forcings influence multiple targets through partially observed effects with different delays and persistence. Direct conditioning leaves the evolution and target-specific influence of these effects implicit in the predictor, while static relational graphs specify connections without tracking evolving effects. This motivates representing future-driv