Latent Twin Operator

DOI10.48550/arXiv.2609.35531MaRDI QIDQ7360741arXivFDO

Authors Deepanshu Verma, Matthias Chung, Riley Chen

Publication date 28 September 2026

Published in arXiv

Tldr We introduce the Latent Twin Operator (LTO), a convolutional Conditional Neural Process for time-evolving PDEs that accepts context sets of sensor observations with variable placement and resolution, enabling arbitrary query resolutions; LTO achieves strong accuracy under one-step comparisons and transfers between native and coarser resolutions.

Full work available at URL https://doi.org/10.48550/arXiv.2609.35531

arXiv classification

cs.CEmath.NA

Cites work

• A Mathematical Guide to Operator Learning
• All of Nonparametric Statistics
• Attention Is All You Need
• Choose a Transformer: Fourier or Galerkin
• Continuum attention for neural operators
• Discretization Error of Fourier Neural Operators
• Fourier Neural Operator for Parametric Partial Differential Equations
• GNOT: A General Neural Operator Transformer for Operator Learning
• Latent neural PDE solver: a reduced-order modeling framework for partial differential equations
• Local Polynomial Modelling and Its Applications
• Message Passing Neural PDE Solvers
• Physics-conforming Latent Twins
• Scattered Data Approximation

This page was built for publication: Latent Twin Operator

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7360741)