Efficient Temporal Tokenization for Mobility Prediction with Large Language Models
This work addresses mobility prediction for urban planning and location-based services, offering incremental efficiency gains through a novel tokenization method.
The paper tackles the problem of mobility prediction by introducing RHYTHM, a framework that uses large language models (LLMs) for spatio-temporal prediction with hierarchical tokenization, resulting in a 2.4% accuracy improvement, 5.0% increase on weekends, and 24.6% reduction in training time compared to state-of-the-art methods.
We introduce RHYTHM (Reasoning with Hierarchical Temporal Tokenization for Human Mobility), a framework that leverages large language models (LLMs) as spatio-temporal predictors and trajectory reasoners. RHYTHM partitions trajectories into daily segments encoded as discrete tokens with hierarchical attention, capturing both daily and weekly dependencies while substantially reducing the sequence length. Token representations are enriched with pre-computed prompt embeddings via a frozen LLM, enhancing the model's ability to capture interdependencies without extensive computational overhead. By freezing the LLM backbone, RHYTHM achieves significant computational efficiency. Evaluation on three real-world datasets demonstrates a 2.4% improvement in accuracy, 5.0% increase on weekends, and 24.6% reduction in training time compared to state-of-the-art methods.