<b>S</b><b>POK: Tokenizing Geographic Space for Enhanced Spatial Reasoning in GeoAI</b>

<br>Tokenization is a fundamental process that feeds data into AI models by breaking data into discrete, learnable units known as tokens. For instance, tokenizing text supports GPT in understanding semantic and contextual relationships within language. However, geospatial tokenization, the process of transforming spatial-temporal data into discrete tokens for geographic modeling, remains significantly underexplored, which has constrained the spatial reasoning capabilities of AI models. This paper presents <b>SPOK</b>, a geo<b>sp</b>atial t<b>ok</b>enization approach that explicitly encodes spatial relationships into spatial tokens. Using urban space as an example, SPOK defines parcels delimited by road links as the basic spatial tokens and employs dynamic location referencing to encode their relative spatial relationships. By embedding these spatial relationships into high-dimensional vectors, SPOK effectively queries and models interactions among spatial tokens in latent space, thereby enhancing spatial reasoning in urban environments. We evaluate SPOK through dynamic mobility flow prediction, and the results demonstrate that SPOK can infer origin-destination patterns without prior spatial interaction data, outperforming baselines with up to a 20% reduction in RMSE and a 14% increase in R². By offering how spatial relationships can be tokenized, this study would like to call for attention on geospatial tokenization to develop geospatial foundation models and GeoAI<br><br><br><b>The SPOK model Content:</b><br><br>This project is linked to a submission to the International Journal of Geographical Information Science, entitled “<b>S</b><b>POK: Tokenizing Geographic Space for Enhanced Spatial Reasoning in GeoAI</b>”, which proposes a geospatial tokenization framework that tokenizes geographic space into discrete tokens and can represent explicit spatial relationships between tokens in a complex geospatial context.<br><br>The project includes the code and synthetic data supporting the SPOK model. Please see ReadMe.md for code implementation details.<br><br>Contact:[email protected]<br>