Sea surface suspended concentrations prediction based on pre-trained transformer
Tao Su, Leilei Wang, Yu Song, Zhongli Bai · 2025
Read Sea surface suspended concentrations prediction based on pre-trained transformer by Tao Su, Leilei Wang, Yu Song, Zhongli Bai (2025) 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/pmc-12749083.txt.
Description
Tao Su, Leilei Wang, Yu Song, and Zhongli Bai forecast sea sediment loads with a borrowed language model in a 2025 paper. Fine sediment hanging in sea surface water shapes ports, habitats, and water quality. Tracking it is hard, since winds, tides, and human acts all stir it. Field and satellite data are sparse, so data-hungry models starve.
Their model starts from a transformer trained on billions of words. Fresh input layers reshape satellite images to fit that model. In fine-tuning, the core of the language model stays frozen while new parts learn the sea task. That way, knowledge from text training carries over. Tests put the model at the top for rebuilding sediment maps, the authors report.
Catalog record
| CREATOR | Tao Su, Leilei Wang, Yu Song, Zhongli Bai |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 38 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |