Unlocking Location Intelligence: A Survey from Deep Learning to The LLM Era

2025-05-13

AI Summary

This study addresses the challenge of transforming geospatial data into actionable knowledge amid the paradigm shift from deep learning to large language models (LLMs). Method: We systematically trace the evolution of geospatial representation learning, propose the first cross-generational taxonomy of location intelligence—spanning data, methodology, and application dimensions—and introduce an LLM-enabled geospatial cross-modal reasoning paradigm with explicit mechanisms for joint modeling of structured and unstructured data. Leveraging CNNs/RNNs/Graph NNs, self-supervised representation learning, multimodal fusion, and geographically grounded LLM fine-tuning and prompt engineering, we construct a dynamic knowledge graph covering 1,000+ scholarly works (open-sourced on GitHub). Contribution/Results: We synthesize seven open challenges and five frontier research directions, delivering a reusable methodological framework for urban computing, intelligent transportation, and related domains.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Location Intelligence (LI), the science of transforming location-centric geospatial data into actionable knowledge, has become a cornerstone of modern spatial decision-making. The rapid evolution of Geospatial Representation Learning is fundamentally reshaping LI development through two successive technological revolutions: the deep learning breakthrough and the emerging large language model (LLM) paradigm. While deep neural networks (DNNs) have demonstrated remarkable success in automated feature extraction from structured geospatial data (e.g., satellite imagery, GPS trajectories), the recent integration of LLMs introduces transformative capabilities for cross-modal geospatial reasoning and unstructured geo-textual data processing. This survey presents a comprehensive review of geospatial representation learning across both technological eras, organizing them into a structured taxonomy based on the complete pipeline comprising: (1) data perspective, (2) methodological perspective and (3) application perspective. We also highlight current advancements, discuss existing limitations, and propose potential future research directions in the LLM era. This work offers a thorough exploration of the field and providing a roadmap for further innovation in LI. The summary of the up-to-date paper list can be found in https://github.com/CityMind-Lab/Awesome-Location-Intelligence and will undergo continuous updates.

Problem

• Surveying deep learning and LLMs in geospatial data analysis
• Exploring cross-modal reasoning for unstructured geo-textual data
• Reviewing advancements and limitations in location intelligence

Innovation

• Deep learning for geospatial feature extraction
• LLMs for cross-modal geospatial reasoning
• Structured taxonomy for geospatial learning pipeline