Geography-Aware Large Language Models for Next POI Recommendation

2025-05-18

AI Summary

To address the challenge of modeling spatial dependencies from sparse GPS coordinates and capturing missing POI transition relationships in large language models (LLMs) for next-POI recommendation, this paper proposes a Geocoding Coordinate Injection Module (GCIM) and a POI Alignment Module (PAM). Our method integrates hierarchical Fourier positional encoding, POI graph-structured embeddings, LLM-based semantic space alignment, and geography-aware prompt enhancement—enabling the first multi-granularity spatial representation learning fused with POI semantic relations. Crucially, it supports zero-shot generalization to unseen POIs. Evaluated on three real-world datasets, our approach achieves state-of-the-art performance, significantly improving recommendation accuracy for long-tail POIs and enhancing cross-regional transferability.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• 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

Abstract

The next Point-of-Interest (POI) recommendation task aims to predict users' next destinations based on their historical movement data and plays a key role in location-based services and personalized applications. Accurate next POI recommendation depends on effectively modeling geographic information and POI transition relations, which are crucial for capturing spatial dependencies and user movement patterns. While Large Language Models (LLMs) exhibit strong capabilities in semantic understanding and contextual reasoning, applying them to spatial tasks like next POI recommendation remains challenging. First, the infrequent nature of specific GPS coordinates makes it difficult for LLMs to model precise spatial contexts. Second, the lack of knowledge about POI transitions limits their ability to capture potential POI-POI relationships. To address these issues, we propose GA-LLM (Geography-Aware Large Language Model), a novel framework that enhances LLMs with two specialized components. The Geographic Coordinate Injection Module (GCIM) transforms GPS coordinates into spatial representations using hierarchical and Fourier-based positional encoding, enabling the model to understand geographic features from multiple perspectives. The POI Alignment Module (PAM) incorporates POI transition relations into the LLM's semantic space, allowing it to infer global POI relationships and generalize to unseen POIs. Experiments on three real-world datasets demonstrate the state-of-the-art performance of GA-LLM.

Problem

• Modeling geographic information for accurate POI recommendations
• Capturing POI transition relations to understand user movements
• Adapting LLMs for spatial tasks with precise GPS contexts

Innovation

• Hierarchical and Fourier-based GPS coordinate encoding
• POI transition relations integration into LLM
• Geography-aware LLM framework for spatial tasks