Mesh adaptivity and Tensor decomposition

[mgenet]

Hello,
We are working at the interface of adaptive finite elements and tensor decomposition, see [Škardová, K., Daby-Seesaram, A., & Genet, M. (2025). Finite element neural network interpolation: Part I—interpretable and adaptive discretization for solving PDEs. Computational Mechanics.], [Daby-Seesaram, A., Škardová, K., & Genet, M. (2025). Finite element neural network interpolation: Part II—hybridisation with the proper generalised decomposition for non-linear surrogate modelling. Computational Mechanics.] and https://github.com/AlexandreDabySeesaram/NeuROM. TensorMesh looks like a really cool framework—do you have any thoughts on how much effort it would be (if feasible at all) to implement mesh adaptivity (local and global) as well as tensor decomposition techniques into it?