analiza aiqbench
This paper uses reverse-geocoded place names as a self-supervised signal to train location encoders, so the representations capture administrative boundaries that existing pretraining objectives, such as imagery or climate variables, leave out. The reported gains are on downstream geospatial tasks, and the authors say code and weights are not yet public, so the claims are not yet independently checkable. For ML readers, the transferable idea is using naturally occurring text as supervision for spatial representations, along with the residual-quantized semantic IDs that make the encodings more interpretable.
Learning Administrative Boundary-Aware Location Encoders via Reverse Geocoding
arXiv:2610.24306v1 Announce Type: new Abstract: Location encoders learn compact representations of latitude--longitude coordinates for downstream geospatial prediction tasks. Existing approaches are pretrained using continuous or naturally occurring supervision, such as imagery, climate variables, or species distributions, and therefore do not explicitly capture administrative boundaries or place names, despite their strong influence on many real-world geospatial phenomena. We introduce, to our knowledge, the first method for learning administrative boundary-aware location encoders, using reverse-geocoded text as a self-supervised training signal. Our approach can be applied to existing location encoder architectures and consistently improves their downstream performance. We further introduce RevGeoRQ, a custom architecture tailored to reverse-geocoding supervision, which yields additional gains and exceeds contemporary location encoders on several downstream tasks. Finally, the use of human-interpretable reverse-geocoded strings and residual-quantized semantic IDs provides a more interpretable representation of geographic location than existing approaches. Code and model weights to be released soon.
Abstract:Location encoders learn compact representations of latitude--longitude coordinates for downstream geospatial prediction tasks. Existing approaches are pretrained using continuous or naturally occurring supervision, such as imagery, climate variables, or species distributions, and therefore do not explicitly capture administrative boundaries or place names, despite their strong influence on many real-world geospatial phenomena. We introduce, to our knowledge, the first method for learning administrative boundary-aware location encoders, using reverse-geocoded text as a self-supervised training signal. Our approach can be applied to existing location encoder architectures and consistently improves their downstream performance. We further introduce RevGeoRQ, a custom architecture tailored to reverse-geocoding supervision, which yields additional gains and exceeds contemporary location encoders on several downstream tasks. Finally, the use of human-interpretable reverse-geocoded strings and residual-quantized semantic IDs provides a more interpretable representation of geographic location than existing approaches. Code and model weights to be released soon.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
arXiv:2610.24306 [cs.LG]
(or arXiv:2610.24306v2 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.24306 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 238 cuvinte