Hybrid GNN-FEM surrogate replaces only the phase-field update step in a staggered incremental scheme for phase-field fracture, using dimensionless mesh features and physics-informed loss to generalize across geometries, loads, materials, and discretizations.
arXiv preprint arXiv:2211.08064 , year=
12 Pith papers cite this work. Polarity classification is still indexing.
12
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
verdicts
UNVERDICTED 12roles
background 2polarities
background 2representative citing papers
A Creator-Inspector multi-agent LLM pipeline for constitutive artificial neural networks increases the rate of models satisfying all nine physical constraints to 100% or 56% depending on the LLM backbone.
A physics-informed Fourier-wavelet transformer model reports the lowest normalized mean-squared error on cylinder-wake and fluid-structure interaction velocity-field benchmarks compared with spectral, transformer, operator-learning, and PINN baselines.
OrthoReg penalizes overlap between symbolic and neural components in hybrid dynamical models, improving symbolic recovery and out-of-distribution behavior under partial library mismatch.
AOT-POT adaptively reshapes complex PDE solution operators via input-dependent transformations and parallel stream mixing to enable effective large-scale pre-training, yielding SOTA results on 12 benchmarks with minimal added parameters.
ST-PT turns transformers into explicit factor graphs for time series, enabling structural injection of symbolic priors, per-sample conditional generation, and principled latent autoregressive forecasting via MFVI iterations.
MH-PINN compactifies unbounded domains with mapping and enforces wave boundary conditions through network architecture for efficient, accurate simulations.
A physics-informed MLP reconstructs high-fidelity 4D spectra from only 1/32 of the samples in experimental 2DIR hyperspectral imaging.
Graph-based summary statistics on pulsar timing residuals detect SGWB down to strain amplitude 1.2e-15 and yield 2.3 sigma evidence in NANOGrav 15-year data via clustering coefficient and edge weight measures.
Introduces Laplace-approximated Bayesian PINNs for automatic loss-weight optimization when solving PDEs such as heat, wave, and Burgers equations.
PINNSur applies PINNs to surface PDEs by neural approximation of normals and operator projection, with an added empirical test for convergence behavior.
A scoping review of physics-informed machine learning for seismic wave propagation finds applications in forward and inverse problems with often comparable accuracy at lower cost, while identifying gaps in benchmarking, training cost, and 3D/experimental validation.
citing papers explorer
-
A Hybrid GNN-FEM Framework for Phase-Field Fracture Simulation. Physics-Preserving Hybridization for Generalizable Surrogate Modeling
Hybrid GNN-FEM surrogate replaces only the phase-field update step in a staggered incremental scheme for phase-field fracture, using dimensionless mesh features and physics-informed loss to generalize across geometries, loads, materials, and discretizations.
-
LLM-driven design of physics-constrained constitutive models: two agents are better than one
A Creator-Inspector multi-agent LLM pipeline for constitutive artificial neural networks increases the rate of models satisfying all nine physical constraints to 100% or 56% depending on the LLM backbone.
-
A Physics-Informed Fourier-Wavelet Transformer for Multiscale Computational Fluid Dynamics Surrogate Modeling
A physics-informed Fourier-wavelet transformer model reports the lowest normalized mean-squared error on cylinder-wake and fluid-structure interaction velocity-field benchmarks compared with spectral, transformer, operator-learning, and PINN baselines.
-
OrthoReg: Orthogonal Regularization for Hybrid Symbolic-Neural Dynamical Systems
OrthoReg penalizes overlap between symbolic and neural components in hybrid dynamical models, improving symbolic recovery and out-of-distribution behavior under partial library mismatch.
-
AOT-POT: Adaptive Operator Transformation for Large-Scale PDE Pre-training
AOT-POT adaptively reshapes complex PDE solution operators via input-dependent transformations and parallel stream mixing to enable effective large-scale pre-training, yielding SOTA results on 12 benchmarks with minimal added parameters.
-
Exploring the Potential of Probabilistic Transformer for Time Series Modeling: A Report on the ST-PT Framework
ST-PT turns transformers into explicit factor graphs for time series, enabling structural injection of symbolic priors, per-sample conditional generation, and principled latent autoregressive forecasting via MFVI iterations.
-
Mapping-based Hard-constrained Physics-Informed Neural Networks for unbounded wave problems
MH-PINN compactifies unbounded domains with mapping and enforces wave boundary conditions through network architecture for efficient, accurate simulations.
-
Accelerating 4D Hyperspectral Imaging through Physics-Informed Neural Representation and Adaptive Sampling
A physics-informed MLP reconstructs high-fidelity 4D spectra from only 1/32 of the samples in experimental 2DIR hyperspectral imaging.
-
Graph-based Summary Statistics for Revealing the Stochastic Gravitational Wave Background in Pulsar Timing Arrays
Graph-based summary statistics on pulsar timing residuals detect SGWB down to strain amplitude 1.2e-15 and yield 2.3 sigma evidence in NANOGrav 15-year data via clustering coefficient and edge weight measures.
-
Bayesian Reasoning for Physics Informed Neural Networks
Introduces Laplace-approximated Bayesian PINNs for automatic loss-weight optimization when solving PDEs such as heat, wave, and Burgers equations.
-
PINNsur: Physics-Informed Neural Networks for PDEs on Curved Surfaces
PINNSur applies PINNs to surface PDEs by neural approximation of normals and operator projection, with an added empirical test for convergence behavior.
-
A Scoping Review of Physics Informed Machine Learning for Wave Propagation Modeling in Seismology
A scoping review of physics-informed machine learning for seismic wave propagation finds applications in forward and inverse problems with often comparable accuracy at lower cost, while identifying gaps in benchmarking, training cost, and 3D/experimental validation.