Why Users Go There: World Knowledge-Augmented Generative Next POI Recommendation

TL;DR

This paper introduces AWARE, a novel LLM-based POI recommendation system that generates personalized, context-aware narratives incorporating regional and temporal knowledge to improve recommendation accuracy.

Contribution

AWARE uniquely integrates external world knowledge with personalized user context through an LLM agent, enhancing POI recommendations with regional and temporal awareness.

Findings

AWARE outperforms baselines with up to 12.4% relative improvement.

The model effectively captures regional cultural traits and seasonal trends.

Personalized narratives improve the relevance of POI recommendations.

Abstract

Generative point-of-interest (POI) recommendation models based on large language models (LLMs) have shown promising results by formulating next POI prediction as a sequence generation task. However, the knowledge encoded in these models remains fixed after training, making them unable to perceive evolving real-world conditions that shape user mobility decisions, such as local events and cultural trends. To bridge this gap, we propose AWARE (Agent-based World knowledge Augmented REcommendation), which employs an LLM agent to generate location- and time-aware contextual narratives that capture regional cultural characteristics, seasonal trends, and ongoing events relevant to each user. Rather than introducing generic or noisy information, AWARE further anchors these narratives in each user's behavioral context, grounding external world knowledge in personalized spatial-temporal patterns. Extensive experiments on three real-world datasets demonstrate that AWARE consistently outperforms competitive baselines, achieving up to 12.4% relative improvement.