TraXion: Rethinking Pre-training Frameworks for Mobility and Beyond

2026-05-07

AI Summary

Existing temporal pretraining approaches often treat human mobility trajectories as simple text sequences, overlooking three core characteristics inherent to multi-entity spatiotemporal event streams (MESES): structured tuple-based events, persistent user-specific traits, and cross-user co-occurrence dependencies. To address this limitation, this work formally introduces three axioms characterizing MESES and proposes TraXion, a unified pretraining framework that jointly models the joint distribution of locations, timestamps, and activities, individual behavioral consistency, and inter-user co-occurrence patterns through purpose-designed self-supervised objectives and neural architectures. Evaluated across six diverse mobility datasets, TraXion significantly outperforms task-specific baselines and achieves state-of-the-art or competitive performance in heterogeneous downstream applications—including enterprise authentication logs and ICU mortality prediction—demonstrating its effectiveness in cross-domain unified modeling.

Abstract

Human mobility differs from text and from generic time series in three structural ways: visits are tuple-valued events whose meaning depends on the joint distribution over location, time, and activity; users carry persistent signatures across trajectories; and visits are not independent across users, since co-location at shared places is a primary signal. Existing pre-training recipes for mobility import objectives from language modeling, treating trajectories as sentences and visits as tokens, an analogy that fails against each of the three properties above. These properties define a broader class, multi-entity spatiotemporal event streams (MESES), spanning enterprise authentication logs, electronic health records, and other event-stream domains where entities share infrastructure, schedules, or contexts. We make the properties precise as three axioms that any pre-training framework for MESES should satisfy, and introduce TraXion, whose objectives and architecture are jointly designed to meet them. A single TraXion checkpoint per dataset beats task-specific baselines on every task across six public mobility datasets covering anomaly detection, next-POI recommendation, next-visit prediction, and social-link prediction. The same recipe, applied unchanged to enterprise authentication logs and ICU mortality prediction, matches or exceeds prior work on both, showing that event streams from domains as different as mobility, security, and healthcare can be modeled under a single framework.

Problem

• human mobility
• pre-training framework
• spatiotemporal event streams
• multi-entity
• MESES

Innovation

• multi-entity spatiotemporal event streams
• pre-training framework
• human mobility modeling
• TraXion
• cross-domain generalization