Coordinates from Context: Using LLMs to Ground Complex Location References
By: Tessa Masis, Brendan O'Connor
Potential Business Impact:
Helps computers find places from messy text.
Geocoding is the task of linking a location reference to an actual geographic location and is essential for many downstream analyses of unstructured text. In this paper, we explore the challenging setting of geocoding compositional location references. Building on recent work demonstrating LLMs' abilities to reason over geospatial data, we evaluate LLMs' geospatial knowledge versus reasoning skills relevant to our task. Based on these insights, we propose an LLM-based strategy for geocoding compositional location references. We show that our approach improves performance for the task and that a relatively small fine-tuned LLM can achieve comparable performance with much larger off-the-shelf models.
Similar Papers
Geospatial Mechanistic Interpretability of Large Language Models
Machine Learning (CS)
Shows how computers "see" maps and places.
SpatialGeo:Boosting Spatial Reasoning in Multimodal LLMs via Geometry-Semantics Fusion
CV and Pattern Recognition
Helps computers understand 3D shapes and where things are.
From Time and Place to Preference: LLM-Driven Geo-Temporal Context in Recommendations
Information Retrieval
Helps movie suggestions understand holidays and seasons.