Reasoning Over Space: Enabling Geographic Reasoning for LLM-Based Generative Next POI Recommendation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911610116767744 |
|---|---|
| author | Lv, Dongyi Ding, Qiuyu Xu, Heng-Da Sun, Zhaoxu Wang, Zhi Xiong, Feng Xu, Mu |
| author_facet | Lv, Dongyi Ding, Qiuyu Xu, Heng-Da Sun, Zhaoxu Wang, Zhi Xiong, Feng Xu, Mu |
| contents | Generative recommendation with large language models (LLMs) reframes prediction as sequence generation, yet existing LLM-based recommenders remain limited in leveraging geographic signals that are crucial in mobility and local-services scenarios. Here, we present Reasoning Over Space (ROS), a framework that utilizes geography as a vital decision variable within the reasoning process. ROS introduces a Hierarchical Spatial Semantic ID (SID) that discretizes coarse-to-fine locality and POI semantics into compositional tokens, and endows LLM with a three-stage Mobility Chain-of-Thought (CoT) paradigm that models user personality, constructs an intent-aligned candidate space, and performs locality informed pruning. We further align the model with real world geography via spatial-guided Reinforcement Learning (RL). Experiments on three widely used location-based social network (LBSN) datasets show that ROS achieves over 10% relative gains in hit rate over strongest LLM-based baselines and improves cross-city transfer, despite using a smaller backbone model. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_04562 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Reasoning Over Space: Enabling Geographic Reasoning for LLM-Based Generative Next POI Recommendation Lv, Dongyi Ding, Qiuyu Xu, Heng-Da Sun, Zhaoxu Wang, Zhi Xiong, Feng Xu, Mu Artificial Intelligence Generative recommendation with large language models (LLMs) reframes prediction as sequence generation, yet existing LLM-based recommenders remain limited in leveraging geographic signals that are crucial in mobility and local-services scenarios. Here, we present Reasoning Over Space (ROS), a framework that utilizes geography as a vital decision variable within the reasoning process. ROS introduces a Hierarchical Spatial Semantic ID (SID) that discretizes coarse-to-fine locality and POI semantics into compositional tokens, and endows LLM with a three-stage Mobility Chain-of-Thought (CoT) paradigm that models user personality, constructs an intent-aligned candidate space, and performs locality informed pruning. We further align the model with real world geography via spatial-guided Reinforcement Learning (RL). Experiments on three widely used location-based social network (LBSN) datasets show that ROS achieves over 10% relative gains in hit rate over strongest LLM-based baselines and improves cross-city transfer, despite using a smaller backbone model. |
| title | Reasoning Over Space: Enabling Geographic Reasoning for LLM-Based Generative Next POI Recommendation |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2601.04562 |