Researchers built a three-dimensional physics-informed neural network to simulate pulsatile blood flow in the human aorta, focusing on abdominal aortic aneurysm haemodynamics. The approach embeds physical laws into the training process, allowing time-resolved simulation without conventional mesh-based solvers. The work is a preprint posted to arXiv.
A new arXiv preprint presents a mesh-based super-resolution method that uses graph transformers to reconstruct multiscale detonation flow data. The authors argue such data-driven reconstruction is useful for subgrid closure modeling, faster spatiotemporal forecasting, compression, and as an upsampling step in simulations. The work appears as a cross-listed revision in the cs.AI and cs.LG categories.
A revised arXiv preprint examines why neural network subgrid stress models perform well in a priori tests but deteriorate in a posteriori large eddy simulations. The authors propose approaches to reduce this discrepancy so that model evaluation better reflects real simulation behavior. The work sits in computational fluid dynamics research rather than commercial AI deployment.