TrajTok: Adaptive Spatial Tokenization for Trajectory Representation Learning

• URL: http://arxiv.org/abs/2605.20134v1
• Date: Tue, 19 May 2026 17:18:32 GMT
• Title: TrajTok: Adaptive Spatial Tokenization for Trajectory Representation Learning
• Abstract summary: We present TrajTok, a trajectory encoder with a simple pretraining recipe for transferable trajectory embeddings.<n>TrajTok first learns a multi-resolution hexagonal cell partition from the spatial distribution of GPS points, converting noisy GPS sequences into discrete cell tokens.<n>It is pretrained with masked-token modeling that recovers both geometric structure and kinematic patterns from partial trajectory observations.
• Score: 14.04282604310535
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Learning generalizable trajectory representations from raw GPS traces remains difficult because the data is continuous, noisy, and irregularly sampled. Spatial tokenization is also challenging: fine grids yield sparse cells with weak embeddings, while coarse grids merge heterogeneous movement patterns into the same token. We present TrajTok, a trajectory encoder with a simple pretraining recipe for transferable trajectory embeddings. TrajTok first learns a multi-resolution hexagonal cell partition from the spatial distribution of GPS points, converting noisy GPS sequences into discrete cell tokens. To capture both geometry and kinematics, it uses a factorized transformer encoder with early per-modality self-attention blocks, cross-attention fusion layers, and spatiotemporal rotary position embeddings, ST-RoPE, to encode where and when each token occurs. TrajTok is pretrained with masked-token modeling that recovers both geometric structure and kinematic patterns from partial trajectory observations. On the Porto dataset, a frozen TrajTok encoder with lightweight task adapters achieves strong performance across trajectory similarity search, classification, estimated time of arrival, and full travel-time regression, outperforming multiple task-specific methods. The same frozen encoder supports both geometry-dominated and kinematics-dominated tasks, suggesting that TrajTok learns transferable trajectory structure rather than task-specific shortcuts. These results indicate that learned multi-resolution spatial tokenization combined with masked-token pretraining is a promising direction for general-purpose trajectory foundation models.

Related papers

• Latent Recurrent Transformer: Architecture Exploration, Training Strategies, and Scaling Behavior [107.2098567818173]
  Latent Recurrent Transformer (LRT) is a lightweight augmentation of autoregressive transformers.<n>LRT reuses a high-level source-layer hidden state from the previous token as recurrent memory for the next token.
  arXiv  Detail & Related papers  (2026-05-26T10:10:26Z)
• Collision-Free Velocity Scheduling for Multi-Agent Systems on Predefined Routes via Inexact-Projection ADMM [0.0]
  In structured multi-agent projects, agents often must follow predefined routes, making rerouting or impossible.<n>This paper addresses route-constrained multi-agent coordination by optimizing waypoint times while preserving each agent's assigned waypoint order and nominal route assignment.
  arXiv  Detail & Related papers  (2026-03-23T12:34:18Z)
• Structural Action Transformer for 3D Dexterous Manipulation [80.07649565189035]
  Cross-embodiment skill transfer is a challenge for high-DoF robotic hands.<n>Existing methods, often relying on 2D observations and temporal-centric action representation, struggle to capture 3D spatial relations.<n>This paper proposes a new 3D dexterous manipulation policy that challenges this paradigm by introducing a structural-centric perspective.
  arXiv  Detail & Related papers  (2026-03-04T11:38:12Z)
• Region-Point Joint Representation for Effective Trajectory Similarity Learning [25.664203648334563]
  textbfRePo is a novel method that encodes textbfRegion-wise and textbfPoint-wise features to capture both spatial context and fine-grained moving patterns.<n>Experiment results show that RePo achieves an average accuracy improvement of 22.2% over SOTA baselines.
  arXiv  Detail & Related papers  (2025-11-17T08:28:18Z)
• Sub-token ViT Embedding via Stochastic Resonance Transformers [51.12001699637727]
  Vision Transformer (ViT) architectures represent images as collections of high-dimensional vectorized tokens, each corresponding to a rectangular non-overlapping patch. We propose a training-free method inspired by "stochastic resonance" The resulting "Stochastic Resonance Transformer" (SRT) retains the rich semantic information of the original representation, but grounds it on a finer-scale spatial domain, partly mitigating the coarse effect of spatial tokenization.
  arXiv  Detail & Related papers  (2023-10-06T01:53:27Z)
• Self-supervised Trajectory Representation Learning with Temporal Regularities and Travel Semantics [30.9735101687326]
  Trajectory Representation Learning (TRL) is a powerful tool for spatial-temporal data analysis and management. Existing TRL works usually treat trajectories as ordinary sequence data, while some important spatial-temporal characteristics, such as temporal regularities and travel semantics, are not fully exploited. We propose a novel Self-supervised trajectory representation learning framework with TemporAl Regularities and Travel semantics, namely START.
  arXiv  Detail & Related papers  (2022-11-17T13:14:47Z)
• CloudAttention: Efficient Multi-Scale Attention Scheme For 3D Point Cloud Learning [81.85951026033787]
  We set transformers in this work and incorporate them into a hierarchical framework for shape classification and part and scene segmentation. We also compute efficient and dynamic global cross attentions by leveraging sampling and grouping at each iteration. The proposed hierarchical model achieves state-of-the-art shape classification in mean accuracy and yields results on par with the previous segmentation methods.
  arXiv  Detail & Related papers  (2022-07-31T21:39:15Z)
• Joint Spatial-Temporal and Appearance Modeling with Transformer for Multiple Object Tracking [59.79252390626194]
  We propose a novel solution named TransSTAM, which leverages Transformer to model both the appearance features of each object and the spatial-temporal relationships among objects. The proposed method is evaluated on multiple public benchmarks including MOT16, MOT17, and MOT20, and it achieves a clear performance improvement in both IDF1 and HOTA.
  arXiv  Detail & Related papers  (2022-05-31T01:19:18Z)
• TransMOT: Spatial-Temporal Graph Transformer for Multiple Object Tracking [74.82415271960315]
  We propose a solution named TransMOT to efficiently model the spatial and temporal interactions among objects in a video. TransMOT is not only more computationally efficient than the traditional Transformer, but it also achieves better tracking accuracy. The proposed method is evaluated on multiple benchmark datasets including MOT15, MOT16, MOT17, and MOT20.
  arXiv  Detail & Related papers  (2021-04-01T01:49:05Z)
• Learning Spatio-Temporal Transformer for Visual Tracking [108.11680070733598]
  We present a new tracking architecture with an encoder-decoder transformer as the key component. The whole method is end-to-end, does not need any postprocessing steps such as cosine window and bounding box smoothing. The proposed tracker achieves state-of-the-art performance on five challenging short-term and long-term benchmarks, while running real-time speed, being 6x faster than Siam R-CNN.
  arXiv  Detail & Related papers  (2021-03-31T15:19:19Z)
• Trajectory saliency detection using consistency-oriented latent codes from a recurrent auto-encoder [0.0]
  Trajectories represent the best way to support progressive dynamic saliency detection. A trajectory will be qualified as salient if it deviates from normal trajectories that share a common motion pattern related to a given context. We show that our method outperforms existing methods on several scenarios drawn from the publicly available dataset of pedestrian trajectories acquired in a railway station.
  arXiv  Detail & Related papers  (2020-12-17T13:29:11Z)
• An Information-Geometric Distance on the Space of Tasks [31.359578768463752]
  This paper prescribes a distance between learning tasks modeled as joint distributions on data and labels. We develop an algorithm to compute the distance which iteratively transports the marginal on the data of the source task to that of the target task. We perform thorough empirical validation and analysis across diverse image classification datasets to show that the coupled transfer distance correlates strongly with the difficulty of fine-tuning.
  arXiv  Detail & Related papers  (2020-11-01T19:48:39Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.