Uni-RS: A Spatially Faithful Unified Understanding and Generation Model for Remote Sensing
Weiyu Zhang, Yuan Hu, Yong Li, Yu Liu · 2026
Read Uni-RS: A Spatially Faithful Unified Understanding and Generation Model for Remote Sensing by Weiyu Zhang, Yuan Hu, Yong Li, Yu Liu (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-2601-17673.txt.
Description
Weiyu Zhang, Yuan Hu, Yong Li, and Yu Liu present Uni-RS for remote sensing. Unified models both understand and generate images. The authors name a flaw they call the spatial reversal curse. Models can describe object locations in images with care. Yet they fail to use the same spatial relations when they generate. In remote sensing, those relations are core meaning.
A map-like image with wrong left-right layout is useless. Uni-RS aims for spatial faithfulness in both directions. Better spatial truth serves land use, disaster, and defense maps. Generation that respects layout can also augment training data. The work names a clear failure mode. Naming it lets the field measure progress on it.
Catalog record
| CREATOR | Weiyu Zhang, Yuan Hu, Yong Li, Yu Liu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-17673