paperGoogle Scholar·25 Oct 2026·K Li, KH Lim

RALLM-STAR: Spatial-Temporal Aligned Retrieval-Augmented LLM for zero-shot next POI recommendation

Retrieval-augmented LLM aligned on space and time for zero-shot next POI recommendation.

recommender-systems Retrieval-Augmented Generation

Depth 2/5Score 2.86

Abstract

Knowledge-Based Systems, 2026 - Elsevier. … While traditional deep learning models suffer from high training costs and poor generalization, Large Language Models (LLMs) offer a promising zero-shot alternative, yet …