Do Sentence Transformers Learn Quasi-Geospatial Concepts from General Text?
Computation and Language
2026-05-07 v1 Machine Learning
Abstract
Sentence transformers are language models designed to perform semantic search. This study investigates the capacity of sentence transformers, fine-tuned on general question-answering datasets for asymmetric semantic search, to associate descriptions of human-generated routes across Great Britain with queries often used to describe hiking experiences. We find that sentence transformers have some zero-shot capabilities to understand quasi-geospatial concepts, such as route types and difficulty, suggesting their potential utility for routing recommendation systems.
Cite
@article{arxiv.2404.04169,
title = {Do Sentence Transformers Learn Quasi-Geospatial Concepts from General Text?},
author = {Ilya Ilyankou and Aldo Lipani and Stefano Cavazzi and Xiaowei Gao and James Haworth},
journal= {arXiv preprint arXiv:2404.04169},
year = {2026}
}
Comments
Presented at the Second International Workshop on Geographic Information Extraction from Texts at ECIR 2024 (https://geo-ext.github.io/GeoExT2024/program/)