Location embeddings from geographic INRs can be decomposed into sparse latent concepts, natural language concepts, and visual features while retaining high reconstruction capability.
Earth Embeddings: Towards AI-Centric Representations of Our Planet
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
MELT and SALT match the two-modality SATCLIP baseline on four downstream tasks but show no consistent gains from extra modalities, indicating the location encoder itself is the bottleneck.
Linear classifier on Clay v1.5 embeddings produces continuous biome probabilities that raise mean per-species AUC for occurrence prediction from 0.570 (discrete labels) to 0.618 on 10,015 Brazilian forest plots.
citing papers explorer
-
What's in an Earth Embedding? An Explainability Analysis of Location Encoders
Location embeddings from geographic INRs can be decomposed into sparse latent concepts, natural language concepts, and visual features while retaining high reconstruction capability.
-
Multi-Modal Contrastive Learning for Implicit Earth Embeddings via Location Tying
MELT and SALT match the two-modality SATCLIP baseline on four downstream tasks but show no consistent gains from extra modalities, indicating the location encoder itself is the bottleneck.