AI RESEARCH
SurF: A Generative Model for Multivariate Irregular Time Series Forecasting
arXiv CS.LG
•
ArXi:2605.14069v1 Announce Type: new Irregularly sampled multivariate event streams remain a stubbornly difficult modality for generative modeling: tokenization-based approaches break down when inter-event intervals vary by orders of magnitude, and neural temporal point processes are bottlenecked by window-level numerical quadrature.