Metraipolis › archive › dc_2bb0f804571f
FLEX-WAM: Flexible Block-Causal World-Action Models for Long-Horizon Imagination and Planning — arXiv:2610.05483
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.05483
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T05:26:57.316Z
- SHA-256 of the archived text
4a2d4ce1771f3fdab0349bdd4b75de4e030039a54c907e65b6f9664764e9116a- Stable URL of this page
https://metraipolis.com/documento/dc_2bb0f804571f
How to cite
FLEX-WAM: Flexible Block-Causal World-Action Models for Long-Horizon Imagination and Planning — arXiv:2610.05483 — arXiv, https://arxiv.org/abs/2610.05483 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2bb0f804571f (Metraipolis archive dc_2bb0f804571f, sha256 4a2d4ce1771f3fdab0349bdd4b75de4e030039a54c907e65b6f9664764e9116a)
BibTeX
@misc{dc2bb0f804571f,
title = {FLEX-WAM: Flexible Block-Causal World-Action Models for Long-Horizon Imagination and Planning — arXiv:2610.05483},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.05483},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2bb0f804571f; sha256 4a2d4ce1771f3fdab0349bdd4b75de4e030039a54c907e65b6f9664764e9116a},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.05483. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
World--action models (WAMs) promise a unified model that predicts action-conditioned futures, generates feasible actions, and supports planning in imagination. However, existing joint video--action models often use computationally heavy, fixed-horizon backbones ill-suited to streaming inference and stable long-horizon open-loop rollouts. We introduce FLEX-WAM, a Flexible and Efficient Block-Causal World--Action Model for unified simulation and policy inference. FLEX-WAM supports variable-length contexts and non-causal prediction horizons, as well as infinite autoregressive generation frame by frame or block by block. Its block-causal, KV-cacheable architecture combines axial attention and blockwise diffusion forcing to enable efficient real-time rollout and deployment-time latency--throughput tradeoffs without retraining. Joint training can nevertheless produce plausible futures that weakly respond to commanded actions. We address this failure mode by balancing state and action flow-matching gradient contributions across the state--action diffusion-noise grid and regulating world-model sampling using Forward-Dynamics (FD) elasticity, an efficient training-time proxy for action responsiveness. Across simulated and real-world datasets, FLEX-WAM achieves superior multi-step prediction quality and latency while producing stable joint state--action rollouts for thousands of steps. As a joint action proposer and simulator within MCTS, it solves long-horizon PushT and all five OGBench Puzzle-4x4 tasks entirely in imagination. On a bimanual OpenArm-based robot, a single checkpoint jointly serves as a play policy and expected-outcome predictor, enabling real-time identification and collection of model--reality mismatches for future self-improvement. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.05483 — 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 2026-10-06T05:24:06.032Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).