arXiv:2606.23453 · v3
Do Location Encoders Capture Spatial Effects? A GeoShapley Benchmark Across Scales
摘要
Location encoders transform geographic coordinates into high dimensional embeddings for downstream machine learning, but it is unclear how well these representations capture interpretable spatial effects. We benchmark whether GeoShapley, a game-theoretic explainer that treats all location features as a single joint player, can recover spatially varying coefficients from models built on location-encoder embeddings. Eleven encoders from the TorchSpatial framework are evaluated against a synthetic process with known coefficients, across three scales (grid, county, global), with and without raw coordinates alongside the embedding, and under untrained and contrastively trained conditions. Measuring recovery as the correlation between estimated and true coefficients, we report how it varies with scale and encoder architecture and compare the embeddings against a raw-coordinate baseline. Recovery of the primary coefficient is consistently high across encoders, whereas recovery of a secondary coefficient is more scale-dependent, differing most at the global scale; the raw-coordinate baseline remains competitive throughout.
作者备注：4 pages, 2 figures. Short paper at ACM SIGSPATIAL 2026. This version updates the tile_ffn results after fixing a TorchSpatial configuration bug that assigned the same embedding to every location in the proceedings experiments. All other results are unchanged