---
title: "DiffVec: Diffusion Model for Trajectory Vector Recovery"
authors:
  - "Jiaqi Duan"
  - "Shengwei Tian"
  - "Long Yu"
  - "Xiangfu Meng"
  - "Ya Zhang"
doi: "10.24963/ijcai.2026/283"
url: "https://ijcai.org/proceedings/2026/283"
pdf: "https://ijcai.org/proceedings/2026/0283.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "2546-2553"
track: "Main Track"
keywords:
  - "Data Mining: Mining spatial and/or temporal data"
  - "Data Mining: Knowledge graphs and knowledge base completion"
  - "Data Mining: Frequent pattern mining"
---

# DiffVec: Diffusion Model for Trajectory Vector Recovery

Jiaqi Duan, Shengwei Tian, Long Yu, Xiangfu Meng, Ya Zhang

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 2546-2553. DOI: [10.24963/ijcai.2026/283](https://doi.org/10.24963/ijcai.2026/283)

## Abstract

The increasing availability of trajectory data is often hampered by sparsity and noise. Existing trajectory recovery methods are further limited by either information loss from coordinate discretization into location IDs, or the inefficiency of conventional diffusion models that require a lengthy denoising process from pure noise. To address these challenges, we propose DiffVec, a novel and efficient diffusion framework for free-space trajectory recovery that operates directly on continuous coordinate data. The backbone of our framework is MVformer, a Transformer-based architecture that models motion vectors—the differences between consecutive coordinates—to better capture the local motion patterns of movement. This model is trained within our ResTraj diffusion paradigm, which commences the denoising process from a structured, interpolated prior to significantly reduce sampling steps. Extensive experiments on the Geolife and Porto datasets demonstrate that DiffVec significantly outperforms state-of-the-art baselines across MAE, MSE, and NDTW.