Feeding an S-DeepONet coarse prediction into a video diffusion model trained on the residual reduces PDE solution errors by 82% on cavity flow and 34% on plastic deformation, beating either single-stage model.
Real-time Inference and Extrapolation via a Diffusion-inspired Temporal Transformer Operator (DiTTO)
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
Extrapolation remains a grand challenge in deep neural networks across all application domains. We propose an operator learning method to solve time-dependent partial differential equations (PDEs) continuously and with extrapolation in time without any temporal discretization. The proposed method, named Diffusion-inspired Temporal Transformer Operator (DiTTO), is inspired by latent diffusion models and their conditioning mechanism, which we use to incorporate the temporal evolution of the PDE, in combination with elements from the transformer architecture to improve its capabilities. Upon training, DiTTO can make inferences in real-time. We demonstrate its extrapolation capability on a climate problem by estimating the temperature around the globe for several years, and also in modeling hypersonic flows around a double-cone. We propose different training strategies involving temporal-bundling and sub-sampling and demonstrate performance improvements for several benchmarks, performing extrapolation for long time intervals as well as zero-shot super-resolution in time.
citation-role summary
citation-polarity summary
fields
cs.CE 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Bridging Sequential Deep Operator Network and Video Diffusion: Residual Refinement of Spatio-Temporal PDE Solutions
Feeding an S-DeepONet coarse prediction into a video diffusion model trained on the residual reduces PDE solution errors by 82% on cavity flow and 34% on plastic deformation, beating either single-stage model.