A Comparative Study of Demonstration Selection for Practical Large Language Models-based Next POI Prediction
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_ | 1866908945130455040 |
|---|---|
| author | Nishida, Ryo Kawarada, Masayuki Ishigaki, Tatsuya Takamura, Hiroya Onishi, Masaki |
| author_facet | Nishida, Ryo Kawarada, Masayuki Ishigaki, Tatsuya Takamura, Hiroya Onishi, Masaki |
| contents | This paper investigates demonstration selection strategies for predicting a user's next point-of-interest (POI) using large language models (LLMs), aiming to accurately forecast a user's subsequent location based on historical check-in data. While in-context learning (ICL) with LLMs has recently gained attention as a promising alternative to traditional supervised approaches, the effectiveness of ICL significantly depends on the selected demonstration. Although previous studies have examined methods such as random selection, embedding-based selection, and task-specific selection, there remains a lack of comprehensive comparative analysis among these strategies. To bridge this gap and clarify the best practices for real-world applications, we comprehensively evaluate existing demonstration selection methods alongside simpler heuristic approaches such as geographical proximity, temporal ordering, and sequential patterns. Extensive experiments conducted on three real-world datasets indicate that these heuristic methods consistently outperform more complex and computationally demanding embedding-based methods, both in terms of computational cost and prediction accuracy. Notably, in certain scenarios, LLMs using demonstrations selected by these simpler heuristic methods even outperform existing fine-tuned models, without requiring further training. Our source code is available at: https://github.com/ryonsd/DS-LLM4POI. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_06207 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | A Comparative Study of Demonstration Selection for Practical Large Language Models-based Next POI Prediction Nishida, Ryo Kawarada, Masayuki Ishigaki, Tatsuya Takamura, Hiroya Onishi, Masaki Computation and Language Artificial Intelligence This paper investigates demonstration selection strategies for predicting a user's next point-of-interest (POI) using large language models (LLMs), aiming to accurately forecast a user's subsequent location based on historical check-in data. While in-context learning (ICL) with LLMs has recently gained attention as a promising alternative to traditional supervised approaches, the effectiveness of ICL significantly depends on the selected demonstration. Although previous studies have examined methods such as random selection, embedding-based selection, and task-specific selection, there remains a lack of comprehensive comparative analysis among these strategies. To bridge this gap and clarify the best practices for real-world applications, we comprehensively evaluate existing demonstration selection methods alongside simpler heuristic approaches such as geographical proximity, temporal ordering, and sequential patterns. Extensive experiments conducted on three real-world datasets indicate that these heuristic methods consistently outperform more complex and computationally demanding embedding-based methods, both in terms of computational cost and prediction accuracy. Notably, in certain scenarios, LLMs using demonstrations selected by these simpler heuristic methods even outperform existing fine-tuned models, without requiring further training. Our source code is available at: https://github.com/ryonsd/DS-LLM4POI. |
| title | A Comparative Study of Demonstration Selection for Practical Large Language Models-based Next POI Prediction |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2604.06207 |