Operator Learning for Reconstructing Flow Fields from Sparse Measurements: a Language Model Approach
Qian Zhang, George Em Karniadakis · 2026
Read Operator Learning for Reconstructing Flow Fields from Sparse Measurements: a Language Model Approach by Qian Zhang, George Em Karniadakis (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-2605-23712.txt.
Description
Zhang and Karniadakis propose operator learning with language model design to rebuild flow fields from sparse measurements, mesh-free. They recast reconstruction as sequence-to-sequence learning, treating sparse measurements as context and unseen spots as queries. The model learns the full flow field from sparse inputs, catching spatial correlations and long-range links. They test it on vortex street simulations, United States temperature data, blood flow simulations, and turbulent jet readings. Scores stay competitive even with under 10 percent seen, showing language models as robust tools for science work.
Catalog record
| CREATOR | Qian Zhang, George Em Karniadakis |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 44378 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.23712