{"ID":2823531,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2601.00452","arxiv_id":"2601.00452","title":"Imitation from Observations with Trajectory-Level Generative Embeddings","abstract":"We consider the offline imitation learning from observations (LfO) where the expert demonstrations are scarce and the available offline suboptimal data are far from the expert behavior. Many existing distribution-matching approaches struggle in this regime because they impose strict support constraints and rely on brittle one-step models, making it hard to extract useful signal from imperfect data. To tackle this challenge, we propose TGE, a trajectory-level generative embedding for offline LfO that constructs a dense, smooth surrogate reward by estimating expert state density in the latent space of a temporal diffusion model trained on offline trajectory data. By leveraging the smooth geometry of the learned diffusion embedding, TGE captures long-horizon temporal dynamics and effectively bridges the gap between disjoint supports, ensuring a robust learning signal even when offline data is distributionally distinct from the expert. Empirically, the proposed approach consistently matches or outperforms prior offline LfO methods across a range of D4RL locomotion and manipulation benchmarks.","short_abstract":"We consider the offline imitation learning from observations (LfO) where the expert demonstrations are scarce and the available offline suboptimal data are far from the expert behavior. Many existing distribution-matching approaches struggle in this regime because they impose strict support constraints and rely on brit...","url_abs":"https://arxiv.org/abs/2601.00452","url_pdf":"https://arxiv.org/pdf/2601.00452v2","authors":"[\"Yongtao Qu\",\"Shangzhe Li\",\"Weitong Zhang\"]","published":"2026-01-01T19:38:37Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[\"Diffusion Model\"]","has_code":false}