A physics-informed Fourier Neural Operator reconstructs particle velocity from DAS strain-rate measurements by enforcing kinematic and elastic-wave-equation constraints, yielding 15.3 dB mean SNR on synthetic tests and low kinematic residuals on real Utah FORGE data without fine-tuning.
International Conference on Learning Representations , year =
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
years
2026 6representative citing papers
CHASM introduces a cross-frequency harmonized axis-separable spectral mixer using a shared channel eigenbasis plus per-frequency positive gains, yielding consistent gains over same-backbone baselines in medical and natural image tasks.
Any convex L-Lipschitz functional on a compact convex subset of a separable Hilbert space can be uniformly approximated to arbitrary accuracy by an explicit convex L-Lipschitz reconstruction from finitely many linear measurements, exactly implementable by a ReLU-MLP.
FNO exhibits strong frequency bias with sharp OOD error growth on high-frequency inputs in wave equations, while DeepONet shows milder degradation despite higher baseline error.
CarCrashNet supplies a large multi-modal crash simulation benchmark and CrashSolver neural model for data-driven full-vehicle crash prediction, validated against experiments and commercial solvers.
citing papers explorer
-
DANTE: Physics-Informed Neural Operator for DAS-to-Velocity Waveform Reconstruction Without Co-located Seismometers
A physics-informed Fourier Neural Operator reconstructs particle velocity from DAS strain-rate measurements by enforcing kinematic and elastic-wave-equation constraints, yielding 15.3 dB mean SNR on synthetic tests and low kinematic residuals on real Utah FORGE data without fine-tuning.
-
CHASM: Cross-frequency Harmonized Axis-Separable Mixing for Spectral Token Operators
CHASM introduces a cross-frequency harmonized axis-separable spectral mixer using a shared channel eigenbasis plus per-frequency positive gains, yielding consistent gains over same-backbone baselines in medical and natural image tasks.
-
Structure-Preserving Reconstruction of Convex Lipschitz Functionals on Hilbert Spaces from Finite Samples
Any convex L-Lipschitz functional on a compact convex subset of a separable Hilbert space can be uniformly approximated to arbitrary accuracy by an explicit convex L-Lipschitz reconstruction from finitely many linear measurements, exactly implementable by a ReLU-MLP.
-
Frequency Bias and OOD Generalization in Neural Operators under a Variable-Coefficient Wave Equation
FNO exhibits strong frequency bias with sharp OOD error growth on high-frequency inputs in wave equations, while DeepONet shows milder degradation despite higher baseline error.
-
CarCrashNet: A Large-Scale Dataset and Hierarchical Neural Solver for Data-Driven Structural Crash Simulation
CarCrashNet supplies a large multi-modal crash simulation benchmark and CrashSolver neural model for data-driven full-vehicle crash prediction, validated against experiments and commercial solvers.
- Mesh Field Theory: Port-Hamiltonian Formulation of Mesh-Based Physics