{"ID":5439450,"CreatedAt":"2026-07-01T01:17:58.482524686Z","UpdatedAt":"2026-07-02T17:27:15.704513176Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2606.30777","arxiv_id":"2606.30777","title":"Unveiling Transferability in Trajectory Prediction via Latent Scene Embeddings","abstract":"The growing availability of trajectory datasets has fueled major advances in data-driven motion prediction. Yet, models trained on one dataset often fail to generalize beyond their training domain as a result of differences in scene layouts, agent behaviors, and sensing conditions. A framework that learns latent representations of datasets and quantifies their similarity using distributional metrics is presented. This large-scale study covers 24 major datasets, including the most widely used motion-prediction benchmarks, and shows that the resulting transferability scores strongly correlate with cross-dataset model performance. The results provide practical guidance for dataset selection, pretraining, and large-scale foundation models for motion prediction, paving the way toward more generalizable and robust predictive systems.","short_abstract":"The growing availability of trajectory datasets has fueled major advances in data-driven motion prediction. Yet, models trained on one dataset often fail to generalize beyond their training domain as a result of differences in scene layouts, agent behaviors, and sensing conditions. A framework that learns latent repres...","url_abs":"https://arxiv.org/abs/2606.30777","url_pdf":"https://arxiv.org/pdf/2606.30777v1","authors":"[\"Theodor Westny\",\"David Axelsson\",\"Björn Olofsson\",\"Erik Frisk\"]","published":"2026-06-29T18:07:34Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","has_code":false}