Bridge augments a graph neural network backbone with time-aware retrieval from a memory of region-time windows to improve cold-start and cross-city urban delivery demand forecasting.
Geolocation representation from large language models are generic enhancers for spatio-temporal learning,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Bridge: Retrieval-Augmented Spatiotemporal Modeling for Urban Delivery Demand
Bridge augments a graph neural network backbone with time-aware retrieval from a memory of region-time windows to improve cold-start and cross-city urban delivery demand forecasting.