Metraipolis › archive › dc_da203d290939

Traj-LeWM: Path-Aware World-Model Planning via Latent Trajectory Cost — arXiv:2608.14125

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.14125
License (as archived)
CC0-1.0
Archived
2026-10-19T02:20:35.835Z
SHA-256 of the archived text
08c172ec3ce1af0ab1f4cd1eace905d6e952046f4d169628c4f12d8435889e80
Stable URL of this page
https://metraipolis.com/documento/dc_da203d290939

How to cite

Traj-LeWM: Path-Aware World-Model Planning via Latent Trajectory Cost — arXiv:2608.14125 — arXiv (en), https://arxiv.org/abs/2608.14125 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_da203d290939 (Metraipolis archive dc_da203d290939, sha256 08c172ec3ce1af0ab1f4cd1eace905d6e952046f4d169628c4f12d8435889e80)
BibTeX
@misc{dcda203d290939,
  title = {Traj-LeWM: Path-Aware World-Model Planning via Latent Trajectory Cost — arXiv:2608.14125},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.14125},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_da203d290939; sha256 08c172ec3ce1af0ab1f4cd1eace905d6e952046f4d169628c4f12d8435889e80},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.14125. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

LeWM is a lightweight visual world model that learns latent dynamics end-to-end from pixels and ranks candidate action sequences by the distance between their predicted endpoints and the goal. However, LeWM has two limitations. First, during training, it learns local next-step transitions without evaluating complete trajectories relative to the task goal. Second, during planning, it ranks candidates solely by predicted endpoint distance. Because model predictions may differ from actual execution outcomes, the candidate whose predicted endpoint is closest to the goal may not perform best when executed in the environment. The evolution of the complete predicted trajectory can therefore provide complementary information beyond endpoint distance. To address these limitations, we propose Traj-LeWM, which retains LeWM's local-dynamics objective and endpoint score while introducing a goal-conditioned latent trajectory cost (LTC) that aggregates trajectory-level information as a complementary signal. During training, LTC-based trajectory-preference supervision complements next-step prediction in shaping the shared representation. During planning, LTC is combined with endpoint distance to incorporate intermediate-path information into candidate ranking. With joint endpoint-plus-LTC scoring, Traj-LeWM outperforms LeWM on Push-T, OGBench-Cube, Reacher, and Two-Room by $3$, $14$, $7$, and $7$ percentage points, respectively. Controlled experiments and ablations further verify the complementary roles of trajectory-level representation shaping and path-aware candidate ranking.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-17 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.14125 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-19T02:02:02.783Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).