Spacetime-GR: A Spacetime-Aware Generative Model for Large Scale Online POI Recommendation

2025-08-22

AI Summary

Point-of-interest (POI) recommendation faces significant challenges in generative modeling due to strong spatiotemporal dynamics in user preferences. Method: We propose the first scalable, spatiotemporally aware generative recommendation model for large-scale online deployment. It introduces a geography-aware hierarchical indexing scheme and a novel spatiotemporal encoding module, integrates multimodal semantic embeddings, and models user actions via sequence modeling under a pretraining-finetuning paradigm. Contribution/Results: Our approach achieves the first online deployment of generative POI recommendation at billion-scale POI volumes; supports flexible multi-format output and downstream task adaptation. Extensive experiments on public and industrial datasets demonstrate state-of-the-art performance—e.g., +12.3% NDCG@10—outperforming existing methods in both accuracy and ranking quality. The model has been successfully deployed in ultra-large-scale production systems.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Natural Language Processing: Generation
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Building upon the strong sequence modeling capability, Generative Recommendation (GR) has gradually assumed a dominant position in the application of recommendation tasks (e.g., video and product recommendation). However, the application of Generative Recommendation in Point-of-Interest (POI) recommendation, where user preferences are significantly affected by spatiotemporal variations, remains a challenging open problem. In this paper, we propose Spacetime-GR, the first spacetime-aware generative model for large-scale online POI recommendation. It extends the strong sequence modeling ability of generative models by incorporating flexible spatiotemporal information encoding. Specifically, we first introduce a geographic-aware hierarchical POI indexing strategy to address the challenge of large vocabulary modeling. Subsequently, a novel spatiotemporal encoding module is introduced to seamlessly incorporate spatiotemporal context into user action sequences, thereby enhancing the model's sensitivity to spatiotemporal variations. Furthermore, we incorporate multimodal POI embeddings to enrich the semantic understanding of each POI. Finally, to facilitate practical deployment, we develop a set of post-training adaptation strategies after sufficient pre-training on action sequences. These strategies enable Spacetime-GR to generate outputs in multiple formats (i.e., embeddings, ranking scores and POI candidates) and support a wide range of downstream application scenarios (i.e., ranking and end-to-end recommendation). We evaluate the proposed model on both public benchmark datasets and large-scale industrial datasets, demonstrating its superior performance over existing methods in terms of POI recommendation accuracy and ranking quality. Furthermore, the model is the first generative model deployed in online POI recommendation services that scale to hundreds of millions of POIs and users.

Problem

• Incorporating spatiotemporal context into generative POI recommendation models
• Addressing large vocabulary challenges in POI recommendation systems
• Enhancing model sensitivity to spatiotemporal variations in user preferences

Innovation

• Geographic-aware hierarchical POI indexing strategy
• Spatiotemporal encoding module for context integration
• Multimodal POI embeddings for semantic enrichment