SPOT converts natural-language scene descriptions into structured OpenStreetMap queries via a fine-tuned LLaMA 3 model and semantic tag bundles, reporting state-of-the-art query-interpretation accuracy on a 195-query journalist benchmark.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
SPOT: Bridging Natural Language and Geospatial Search for Investigative Journalists
SPOT converts natural-language scene descriptions into structured OpenStreetMap queries via a fine-tuned LLaMA 3 model and semantic tag bundles, reporting state-of-the-art query-interpretation accuracy on a 195-query journalist benchmark.