ADS-POI: Agentic Spatiotemporal State Decomposition for Next Point-of-Interest Recommendation

TL;DR

ADS-POI introduces a novel multi-state decomposition approach for next POI recommendation, capturing diverse user behaviors evolving at different spatiotemporal scales to improve prediction accuracy.

Contribution

It proposes a framework that models user behavior with multiple parallel latent states, each with distinct dynamics, enhancing flexibility and robustness over existing methods.

Findings

ADS-POI outperforms state-of-the-art baselines on benchmark datasets.

Decomposing user behavior into multiple states improves recommendation effectiveness.

The model adapts to diverse behavioral patterns through spatiotemporal state evolution.

Abstract

Next point-of-interest (POI) recommendation requires modeling user mobility as a spatiotemporal sequence, where different behavioral factors may evolve at different temporal and spatial scales. Most existing methods compress a user's history into a single latent representation, which tends to entangle heterogeneous signals such as routine mobility patterns, short-term intent, and temporal regularities. This entanglement limits the flexibility of state evolution and reduces the model's ability to adapt to diverse decision contexts. We propose ADS-POI, a spatiotemporal state decomposition framework for next POI recommendation. ADS-POI represents a user with multiple parallel evolving latent sub-states, each governed by its own spatiotemporal transition dynamics. These sub-states are selectively aggregated through a context-conditioned mechanism to form the decision state used for prediction. This design enables different behavioral components to evolve at different rates while remaining coordinated under the current spatiotemporal context. Extensive experiments on three real-world benchmark datasets from Foursquare and Gowalla demonstrate that ADS-POI consistently outperforms strong state-of-the-art baselines under a full-ranking evaluation protocol. The results show that decomposing user behavior into multiple spatiotemporally aware states leads to more effective and robust next POI recommendation. Our code is available at https://github.com/YuZhenyuLindy/ADS-POI.git.