SNAP-FM accelerates nonlinear constraint projection in Physics-Constrained Flow Matching by exploiting block-sparse Jacobian and KKT structures with ExaModels.jl, MadNLP.jl, and GPU sparse factorization on PDE benchmarks.
and Ma, Ruijun and Mahoney, Michael W
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Probabilistic neural network framework embeds linear equality constraints for dynamic chemical process modeling, showing improved accuracy, calibration, and constraint adherence on reduced data plus faster training on large data.
citing papers explorer
-
SNAP-FM: Sparse Nonlinear Accelerated Projection for Physics-Constrained Generative Modeling
SNAP-FM accelerates nonlinear constraint projection in Physics-Constrained Flow Matching by exploiting block-sparse Jacobian and KKT structures with ExaModels.jl, MadNLP.jl, and GPU sparse factorization on PDE benchmarks.
-
Embedding Linear Equality Constraints in Probabilistic Neural Networks for Dynamic Modelling
Probabilistic neural network framework embeds linear equality constraints for dynamic chemical process modeling, showing improved accuracy, calibration, and constraint adherence on reduced data plus faster training on large data.