Metraipolis › archive › dc_eea8092f39ec
ForgeWM: Progressive Causal Training for Few-Step Action-Conditioned Video World Models — arXiv:2608.14022
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.14022
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T02:21:15.727Z
- SHA-256 of the archived text
ce06797f01ecd20f82cf96716b781209f2ddd060005a4c604ccfe1d06437dd3b- Stable URL of this page
https://metraipolis.com/documento/dc_eea8092f39ec
How to cite
ForgeWM: Progressive Causal Training for Few-Step Action-Conditioned Video World Models — arXiv:2608.14022 — arXiv (en), https://arxiv.org/abs/2608.14022 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_eea8092f39ec (Metraipolis archive dc_eea8092f39ec, sha256 ce06797f01ecd20f82cf96716b781209f2ddd060005a4c604ccfe1d06437dd3b)
BibTeX
@misc{dceea8092f39ec,
title = {ForgeWM: Progressive Causal Training for Few-Step Action-Conditioned Video World Models — arXiv:2608.14022},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.14022},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_eea8092f39ec; sha256 ce06797f01ecd20f82cf96716b781209f2ddd060005a4c604ccfe1d06437dd3b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.14022. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Action-conditioned video world models require low-latency causal generation and reliable responses to game-native controls. Although causal distillation enables one- or few-step video synthesis, extending it to interactive world models remains challenging, as discrete keyboard states and continuous mouse motion must remain aligned with temporally compressed latent chunks during causal training and autoregressive rollout. We introduce ForgeWM, a progressive framework that transforms a bidirectional action-conditioned video generator into efficient few-step world models through domain adaptation, teacher-forced causal training, causal consistency distillation, and on-policy distribution matching with a bidirectional teacher. The resulting budget-specialized students operate at steady-state denoising budgets of 1, 2, and 4 steps. ForgeWM further supports a dual-path deployment protocol combining latency-critical interaction with optional replay-time refinement, where the one-step student re-noises and refines its saved draft. On paired Minecraft trajectories, ForgeWM leads the evaluated systems in Imaging Quality, reference-aligned motion-profile agreement, action-sign accuracy, and mouse-control accuracy, while achieving the lowest reference LPIPS; the same four-stage recipe transfers to gamepad-controlled FPS gameplay. Replay-time refinement matches four-step reference quality while remaining roughly three times closer to the experienced trajectory than regeneration from noise. These results demonstrate ForgeWM's effectiveness for controllable few-step video generation. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-17 in cs.AI). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.14022 — 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 2026-10-19T02:02:02.894Z 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).