Metraipolis › archive › dc_72353aa38796
Metro-WM: Long-Horizon Latent Planning with Realisable Sub-Goals — arXiv:2609.31868
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.31868
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T20:54:56.456Z
- SHA-256 of the archived text
308e7e81e9f9c0fb0b84c80a7576ad37757c85fdc5a203c07060f7649d9a5761- Stable URL of this page
https://metraipolis.com/documento/dc_72353aa38796
How to cite
Metro-WM: Long-Horizon Latent Planning with Realisable Sub-Goals — arXiv:2609.31868 — arXiv (en), https://arxiv.org/abs/2609.31868 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_72353aa38796 (Metraipolis archive dc_72353aa38796, sha256 308e7e81e9f9c0fb0b84c80a7576ad37757c85fdc5a203c07060f7649d9a5761)
BibTeX
@misc{dc72353aa38796,
title = {Metro-WM: Long-Horizon Latent Planning with Realisable Sub-Goals — arXiv:2609.31868},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.31868},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_72353aa38796; sha256 308e7e81e9f9c0fb0b84c80a7576ad37757c85fdc5a203c07060f7649d9a5761},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.31868. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Model-predictive control with Joint-Embedding Predictive Architectures (JEPAs) provides a strong zero-shot goal-reaching planner, but it is only effective over short planning horizons. Hierarchical extensions attempt to bridge this gap by learning a macro planner to predict intermediate latent sub-goals to guide the micro planner. In this work, we demonstrate that unconstrained latent sub-goal prediction is fundamentally flawed. A rigorous evaluation reveals that a leading state-of-the-art macro planner routinely emits physically unrealisable sub-goals. To resolve this, we introduce Metro-WM, a hierarchical framework that issues sub-goals by retrieving genuine states from prior experience rather than generating ungrounded latent vectors. Specifically, Metro-WM constructs a graph whose vertices are observed frames from offline expert demonstrations or random-action trajectories, allowing frames from different episodes to be connected and stitched into routes to the goal. Planning over the full graph also makes the system highly robust to execution errors: if the micro planner drifts off course, Metro-WM instantly finds a new optimal path from the current state. Our experiments show that Metro-WM achieves superior long-horizon success rates of up to 37.33 percentage points over the next best hierarchical approach while being up to 10.9 times faster, requiring both 13-56 times less offline compute and fewer tuned hyperparameters. Additional analysis reveals that Metro-WM finds shorter paths than the offline demonstrations, outperforms an oracle relying on the query's own demonstration, and maintains robust performance under extremely sparse dataset conditions. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI) ; Robotics (cs.RO). Pagina del lavoro: https://arxiv.org/abs/2609.31868 — 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 2026-10-07T20:29:00.649Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).