Towards A Foundation Model For Trajectory Intelligence

TL;DR

This paper introduces a large-scale trajectory foundation model trained on extensive real-world check-in data, employing a novel spatial tokenization method to improve trajectory understanding and downstream task performance.

Contribution

The work presents a pre-train and fine-tune paradigm for trajectory modeling, including a novel spatial tokenization block to handle noisy data and large vocabularies.

Findings

Effective learning of trajectory patterns from 2 billion check-ins

Successful fine-tuning on three downstream tasks

Demonstrated potential for trajectory intelligence applications

Abstract

We present the results of training a large trajectory model using real-world user check-in data. Our approach follows a pre-train and fine-tune paradigm, where a base model is pre-trained via masked trajectory modeling and then adapted through fine-tuning for various downstream tasks. To address challenges posed by noisy data and large spatial vocabularies, we propose a novel spatial tokenization block. Our empirical analysis utilizes a comprehensive dataset of over 2 billion check-ins generated by more than 6 million users. Through fine-tuning on 3 downstream tasks we demonstrate that our base model has effectively learned valuable underlying patterns in raw data, enabling its application in meaningful trajectory intelligence tasks. Despite some limitations, we believe this work represents an important step forward in the realization of a foundation model for trajectory intelligence.