Researchers have developed a new type of neural operator called the Spatiotemporal Conditional Operator (STCO) that can make predictions about time-dependent physical systems governed by partial differential equations. Unlike previous approaches, STCO takes into account prescribed conditions such as body motion, inflow, or forcing, which are not determined solely by the observed state. The authors evaluate their method on a benchmark problem involving computational fluid dynamics and show that it reduces errors in field predictions and load calculations compared to existing methods. --- Why it matters: This work is significant for researchers working with time-dependent physical systems because it provides a more accurate way to make predictions about complex phenomena such as fluid flow. By incorporating prescribed conditions, STCO can improve the accuracy of simulations and models used in fields like engineering and climate science. Source: https://arxiv.org/abs/2608.20477