TrajDLM: Topology-Aware Block Diffusion Language Model for Trajectory Generation

TL;DR

TrajDLM is a novel topology-aware block diffusion model for generating realistic GPS trajectories efficiently, outperforming prior methods in speed and accuracy across multiple datasets.

Contribution

It introduces a block diffusion framework that models trajectories as sequences of road segments with topology-aware embeddings, enabling fast and faithful trajectory generation.

Findings

TrajDLM achieves up to 2.8x faster generation than previous models.

It maintains high local similarity metrics across three city datasets.

The model demonstrates strong zero-shot transfer to unseen transportation modes.

Abstract

Generating high-fidelity synthetic GPS trajectories is increasingly important for applications in transportation, urban planning, and what-if scenario simulation, especially as privacy concerns limit access to real-world mobility data. Existing trajectory generation models face a trade-off between efficiency and faithfulness to road network topology: continuous-space methods enable fast generation but ignore the road network, while topology-aware approaches rely on search-based autoregressive decoding that limits generation speed. We propose TrajDLM, a topology-aware trajectory generation framework based on block diffusion language models that bridges this gap. TrajDLM models trajectories as sequences of discrete road segments, combining a block diffusion backbone for efficient denoising, topology-aware embeddings from a road network encoder, and topology-constrained sampling to ensure coherent and realistic trajectories. Across three city-scale datasets, TrajDLM achieves strong performance on fine-grained local similarity metrics while being up to $2.8\times$ faster than prior work, and demonstrates strong zero-shot transfer across domains, including unseen transportation modes. These results highlight the effectiveness of block-wise discrete diffusion as a scalable approach to accurate and efficient trajectory generation. Our code is available at https://github.com/cruiseresearchgroup/TrajDLM/