MEEC equips point clouds with a discrete exterior calculus that satisfies exact conservation and is differentiable in point positions, allowing a single trained kernel to produce compatible physics on unseen geometries and parameters.
hub Canonical reference
Physics-informed machine learning.Nature Reviews Physics, 3(6):422–440
Canonical reference. 100% of citing Pith papers cite this work as background.
16
Pith papers citing it
Background
100% of classified citations
hub tools
citation-role summary
background 6
citation-polarity summary
years
2026 16verdicts
UNVERDICTED 16roles
background 6polarities
background 6representative citing papers
A two-stage contrastive teacher-student framework learns and then projects latent dynamics onto port-Hamiltonian submanifolds from partial observations.
OceanCBM is the first concept bottleneck model for spatiotemporal ocean prediction that uses mixed supervision on physical concepts and a free concept to deliver consistent mechanistic representations for mixed layer heat content forecasts.
Local neural operators on 3x3x3 patches, composed via Schwarz iteration, solve large-scale nonlinear elasticity on arbitrary geometries without domain-specific retraining.
PINN gradient conflicts occur in distinct regimes (persistent directional, magnitude imbalance, or low/transient) that each favor different fixes, with per-loss adapters plus reweighting improving results on forward and multi-physics problems.
Update direction selection for PINN training is cast as a Chebyshev-center problem in the dual cone, yielding an efficient dual formulation with nonconvex convergence guarantees and automatic recovery of scale robustness and simultaneous descent.
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.
NeTMY neural fields with annealed encoding, multiscale optimization, and spectrum-fidelity losses achieve superior localization and distributional accuracy in NV-center inverse sensing by using a tensor power-summed dipolar operator that exposes and mitigates center-collapse failures.
LaWM induces latent transitions from a learned discrete variational principle rather than an unconstrained neural predictor, yielding improved physical consistency on synthetic dynamics and robot benchmarks.
FluidFlow uses conditional flow-matching with U-Net and DiT architectures to predict pressure and friction coefficients on airfoils and 3D aircraft meshes, outperforming MLP baselines with better generalization.
Evolving lifted data vectors under a chaotic dynamical system before softmax classification accelerates training and improves accuracy over standard and lifted-only baselines on perturbed orthogonal vectors.
DLDMF disentangles latent dynamics for parameterized PDEs by feeding parameters into a latent embedding that initializes a parameter-conditioned Neural ODE, then uses dynamic manifold fusion with a shared decoder to reconstruct spatiotemporal fields for better generalization and extrapolation.
The Neural Basis Method uses a predefined neural basis space and operator residual metric to deliver accurate single solves and fast parametric learning for multiscale Darcian dynamics.
AdamFLIP treats PDE constraint residuals in PINNs as a controlled dynamical system, computes Lagrange multipliers via feedback linearization to drive residuals to zero, and applies Adam-style adaptation to the resulting gradient for scalable hard-constrained training.
Transfer learning with pretraining-fine-tuning improves neural parameter estimation accuracy for building RC models by 18.6-49.4% over baselines while removing the need for initial parameter guesses.
AI4S-SDS uses sparse MCTS and differentiable physics alignment to generate valid solvent mixtures and identifies a competitive photoresist developer formulation.
citing papers explorer
-
A meshfree exterior calculus for generalizable and data-efficient learning of physics from point clouds
MEEC equips point clouds with a discrete exterior calculus that satisfies exact conservation and is differentiable in point positions, allowing a single trained kernel to produce compatible physics on unseen geometries and parameters.
-
Identify Then Project: Contrastive Learning of Latent Dynamics from Partial Observations with Port-Hamiltonian Structure
A two-stage contrastive teacher-student framework learns and then projects latent dynamics onto port-Hamiltonian submanifolds from partial observations.
-
OceanCBM: A Concept Bottleneck Model for Mechanistic Interpretability in Ocean Forecasting
OceanCBM is the first concept bottleneck model for spatiotemporal ocean prediction that uses mixed supervision on physical concepts and a free concept to deliver consistent mechanistic representations for mixed layer heat content forecasts.
-
Neural-Schwarz Tiling for Geometry-Universal PDE Solving at Scale
Local neural operators on 3x3x3 patches, composed via Schwarz iteration, solve large-scale nonlinear elasticity on arbitrary geometries without domain-specific retraining.
-
Per-Loss Adapters for Gradient Conflict in Physics-Informed Neural Networks
PINN gradient conflicts occur in distinct regimes (persistent directional, magnitude imbalance, or low/transient) that each favor different fixes, with per-loss adapters plus reweighting improving results on forward and multi-physics problems.
-
Chebyshev Center-Based Direction Selection for Multi-Objective Optimization and Training PINNs
Update direction selection for PINN training is cast as a Chebyshev-center problem in the dual cone, yielding an efficient dual formulation with nonconvex convergence guarantees and automatic recovery of scale robustness and simultaneous descent.
-
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.
-
Neural Fields for NV-Center Inverse Sensing
NeTMY neural fields with annealed encoding, multiscale optimization, and spectrum-fidelity losses achieve superior localization and distributional accuracy in NV-center inverse sensing by using a tensor power-summed dipolar operator that exposes and mitigates center-collapse failures.
-
LaWM: Least Action World Models for Long-Horizon Physical Consistency from Visual Observations
LaWM induces latent transitions from a learned discrete variational principle rather than an unconstrained neural predictor, yielding improved physical consistency on synthetic dynamics and robot benchmarks.
-
FluidFlow: a flow-matching generative model for fluid dynamics surrogates on unstructured meshes
FluidFlow uses conditional flow-matching with U-Net and DiT architectures to predict pressure and friction coefficients on airfoils and 3D aircraft meshes, outperforming MLP baselines with better generalization.
-
Enhancing classification accuracy through chaos
Evolving lifted data vectors under a chaotic dynamical system before softmax classification accelerates training and improves accuracy over standard and lifted-only baselines on perturbed orthogonal vectors.
-
Disentangled Latent Dynamics Manifold Fusion for Solving Parameterized PDEs
DLDMF disentangles latent dynamics for parameterized PDEs by feeding parameters into a latent embedding that initializes a parameter-conditioned Neural ODE, then uses dynamic manifold fusion with a shared decoder to reconstruct spatiotemporal fields for better generalization and extrapolation.
-
Solving and learning advective multiscale Darcian dynamics with the Neural Basis Method
The Neural Basis Method uses a predefined neural basis space and operator residual metric to deliver accurate single solves and fast parametric learning for multiscale Darcian dynamics.
-
AdamFLIP: Adaptive Momentum Feedback Linearization Optimization for Hard Constrained PINN Training
AdamFLIP treats PDE constraint residuals in PINNs as a controlled dynamical system, computes Lagrange multipliers via feedback linearization to drive residuals to zero, and applies Adam-style adaptation to the resulting gradient for scalable hard-constrained training.
-
Transfer Learning for Neural Parameter Estimation applied to Building RC Models
Transfer learning with pretraining-fine-tuning improves neural parameter estimation accuracy for building RC models by 18.6-49.4% over baselines while removing the need for initial parameter guesses.
-
AI4S-SDS: A Neuro-Symbolic Solvent Design System via Sparse MCTS and Differentiable Physics Alignment
AI4S-SDS uses sparse MCTS and differentiable physics alignment to generate valid solvent mixtures and identifies a competitive photoresist developer formulation.