Mag-Mamba: Modeling Coupled spatiotemporal Asymmetry for POI Recommendation

TL;DR

Mag-Mamba introduces a novel complex domain approach to model dynamic spatiotemporal asymmetry in POI recommendation, significantly improving prediction accuracy by capturing time-varying directional mobility patterns.

Contribution

It proposes a phase-driven rotational dynamics framework with a time-conditioned Magnetic Phase Encoder and a complex-valued Mamba module for better modeling of urban mobility.

Findings

Achieves significant performance improvements over state-of-the-art methods.

Effectively models time-varying spatial directionality.

Demonstrates robustness across multiple real-world datasets.

Abstract

Next Point-of-Interest (POI) recommendation is a critical task in location-based services, yet it faces the fundamental challenge of coupled spatiotemporal asymmetry inherent in urban mobility. Specifically, transition intents between locations exhibit high asymmetry and are dynamically conditioned on time. Existing methods, typically built on graph or sequence backbones, rely on symmetric operators or real-valued aggregations, struggling to unify the modeling of time-varying global directionality. To address this limitation, we propose Mag-Mamba, a framework whose core insight lies in modeling spatiotemporal asymmetry as phase-driven rotational dynamics in the complex domain. Based on this, we first devise a Time-conditioned Magnetic Phase Encoder that constructs a time-conditioned Magnetic Laplacian on the geographic adjacency graph, utilizing edge phase differences to characterize the globally evolving spatial directionality. Subsequently, we introduce a Complex-valued Mamba module that generalizes traditional scalar state decay into joint decay-rotation dynamics, explicitly modulated by both time intervals and magnetic geographic priors. Extensive experiments on three real-world datasets demonstrate that Mag-Mamba achieves significant performance improvements over state-of-the-art baselines.