LLM-Aligned Geographic Item Tokenization for Local-Life Recommendation

TL;DR

This paper introduces LGSID, a novel framework that enhances local-life recommendation by aligning large language models with geographic information through reinforcement learning and hierarchical tokenization.

Contribution

It proposes a new RL-based alignment method and hierarchical geographic tokenization to better encode spatial relationships in LLMs for recommendation tasks.

Findings

LGSID outperforms existing models on industry datasets.

The RL-based alignment captures real-world spatial relationships.

Hierarchical tokenization improves geographic representation accuracy.

Abstract

Recent advances in Large Language Models (LLMs) have enhanced text-based recommendation by enriching traditional ID-based methods with semantic generalization capabilities. Text-based methods typically encode item textual information via prompt design and generate discrete semantic IDs through item tokenization. However, in domain-specific tasks such as local-life services, simply injecting location information into prompts fails to capture fine-grained spatial characteristics and real-world distance awareness among items. To address this, we propose LGSID, an LLM-Aligned Geographic Item Tokenization Framework for Local-life Recommendation. This framework consists of two key components: (1) RL-based Geographic LLM Alignment, and (2) Hierarchical Geographic Item Tokenization. In the RL-based alignment module, we initially train a list-wise reward model to capture real-world spatial relationships among items. We then introduce a novel G-DPO algorithm that uses pre-trained reward model to inject generalized spatial knowledge and collaborative signals into LLMs while preserving their semantic understanding. Furthermore, we propose a hierarchical geographic item tokenization strategy, where primary tokens are derived from discrete spatial and content attributes, and residual tokens are refined using the aligned LLM's geographic representation vectors. Extensive experiments on real-world Kuaishou industry datasets show that LGSID consistently outperforms state-of-the-art discriminative and generative recommendation models. Ablation studies, visualizations, and case studies further validate its effectiveness.