Arbitrarily Conditioned Hierarchical Flows for Spatiotemporal Events

ArXi:2605.01226v1 Announce Type: new Events in spatiotemporal systems are ubiquitous, yet modeling their complex distributions remains challenging. Existing point process models often rely on strong structural assumptions and are typically limited to autoregressive, event-by-event prediction. As a result, they struggle to broader inference tasks such as inverse inference, trajectory reconstruction, and recovery of missing event locations. We