Multimodal embeddings from satellite images and road graphs raise accident prediction AUROC to 90.1 percent and attribute 24 percent higher rates to increased precipitation after confounder adjustment.
Multimodal learning with graphs
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Learning Multimodal Embeddings for Traffic Accident Prediction and Causal Estimation
Multimodal embeddings from satellite images and road graphs raise accident prediction AUROC to 90.1 percent and attribute 24 percent higher rates to increased precipitation after confounder adjustment.