HiT-JEPA: A Hierarchical Self-supervised Trajectory Embedding Framework for Similarity Computation

2025/06/17 by Lihuan Li, Li, Lihuan, Hao Xue +7 · 1 citation
Computer Science · Engineering · Social Sciences · #Artificial Intelligence (cs.AI) #Automated Road and Building Extraction #Computer Vision and Pattern Recognition (cs.CV) #Data Management and Algorithms #FOS: Computer and information sciences #Human Mobility and Location-Based Analysis #Machine Learning (cs.LG)

paper · pdf · doi:10.48550/arxiv.2507.00028

openalex publication_date 2025/06/17 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/01

Abstract

The representation of urban trajectory data plays a critical role in effectively analyzing spatial movement patterns. Despite considerable progress, the challenge of designing trajectory representations that can capture diverse and complementary information remains an open research problem. Existing methods struggle in incorporating trajectory fine-grained details and high-level summary in a single model, limiting their ability to attend to both long-term dependencies while preserving local nuances. To address this, we propose HiT-JEPA (Hierarchical Interactions of Trajectory Semantics via a Joint Embedding Predictive Architecture), a unified framework for learning multi-scale urban trajectory representations across semantic abstraction levels. HiT-JEPA adopts a three-layer hierarchy that progressively captures point-level fine-grained details, intermediate patterns, and high-level trajectory abstractions, enabling the model to integrate both local dynamics and global semantics in one coherent structure. Extensive experiments on multiple real-world datasets for trajectory similarity computation show that HiT-JEPA's hierarchical design yields richer, multi-scale representations. Code is available at: https://anonymous.4open.science/r/HiT-JEPA.

Citations

• Blurred Encoding for Trajectory Representation Learning
• UniTraj: Learning a Universal Trajectory Foundation Model from Billion-Scale Worldwide Traces
• Context-Enhanced Multi-View Trajectory Representation Learning: Bridging the Gap through Self-Supervised Models
• SIMformer: Single-Layer Vanilla Transformer Can Learn Free-Space Trajectory Similarity
• T-JEPA: A Joint-Embedding Predictive Architecture for Trajectory Similarity Computation
• Understanding Masked Autoencoders via Hierarchical Latent Variable Models
• Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture
• Contrastive Trajectory Similarity Learning with Dual-Feature Attention
• Pre-training General Trajectory Embeddings with Maximum Multi-view Entropy Coding
• Scaling Vision Transformers to Gigapixel Images via Hierarchical Self-Supervised Learning
• Hierarchical Self-supervised Representation Learning for Movie Understanding
• HiCLRE: A Hierarchical Contrastive Learning Framework for Distantly Supervised Relation Extraction
• Masked Autoencoders Are Scalable Vision Learners
• VICReg: Variance-Invariance-Covariance Regularization for\n Self-Supervised Learning
• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• Adaptive Graph Convolutional Recurrent Network for Traffic Forecasting
• HIBERT: Document Level Pre-training of Hierarchical Bidirectional Transformers for Document Summarization
• Attention Is All You Need
• node2vec: Scalable Feature Learning for Networks
• Efficient Estimation of Word Representations in Vector Space
• Long Short-Term Memory
• BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Cited by

• Learn from your own latents and not from tokens: A sample-complexity theory

Related

• T-JEPA: A Joint-Embedding Predictive Architecture for Trajectory Similarity Computation
• MovSemCL: Movement-Semantics Contrastive Learning for Trajectory Similarity (Extension)
• TrajDiff: Diffusion Bridge Network with Semantic Alignment for Trajectory Similarity Computation
• Region-Point Joint Representation for Effective Trajectory Similarity Learning
• Trajectory Similarity Measurement: An Efficiency Perspective