G-PARC: Graph-Physics Aware Recurrent Convolutional Neural Networks for Spatiotemporal Dynamics on Unstructured Meshes
Jack T. Beerman, Tyler J. Abele, Mehdi Taghizadeh, Andrew Davis, Zoë J. Gray, Negin Alemazkoor, Xinfeng Gao, H.S. Udaykumar, Stephen S. Baek · 2026
Read G-PARC: Graph-Physics Aware Recurrent Convolutional Neural Networks for Spatiotemporal Dynamics on Unstructured Meshes by Jack T. Beerman, Tyler J. Abele, Mehdi Taghizadeh, Andrew Davis, Zoë J. Gray, Negin Alemazkoor, Xinfeng Gao, H.S. Udaykumar, Stephen S. Baek (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-16533.txt.
Description
Beerman and colleagues present G-PARC, a graph-based looping convolutional network for motion on uneven meshes. Pixel-based earlier models were stuck on Cartesian grids, while graph methods struggled in extreme wild regimes. G-PARC finds slopes on uneven graphs with moving least squares kernels and embeds slopes of the governing laws into the network. It beat graph rivals with far fewer settings, spread across uneven space and time grids, and handled moving meshes for bending. It led on benchmarks spanning river flow, shock waves and stretchy solids.
Catalog record
| CREATOR | Jack T. Beerman, Tyler J. Abele, Mehdi Taghizadeh, Andrew Davis, Zoë J. Gray, Negin Alemazkoor, Xinfeng Gao, H.S. Udaykumar, Stephen S. Baek |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 62680 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.16533