Metraipolis › archive › dc_9354fdaeada9
AlayaWorld: Interactive Long-Horizon World Modeling - Full Technical Report (v1.1) — arXiv:2608.13492
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.13492
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T04:17:44.747Z
- SHA-256 of the archived text
0ce69700e301dfbbd7d27b5f9a426d4266e0a5a3b313430baa8eb1561e38b463- Stable URL of this page
https://metraipolis.com/documento/dc_9354fdaeada9
How to cite
AlayaWorld: Interactive Long-Horizon World Modeling - Full Technical Report (v1.1) — arXiv:2608.13492 — arXiv (en), https://arxiv.org/abs/2608.13492 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9354fdaeada9 (Metraipolis archive dc_9354fdaeada9, sha256 0ce69700e301dfbbd7d27b5f9a426d4266e0a5a3b313430baa8eb1561e38b463)
BibTeX
@misc{dc9354fdaeada9,
title = {AlayaWorld: Interactive Long-Horizon World Modeling - Full Technical Report (v1.1) — arXiv:2608.13492},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.13492},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9354fdaeada9; sha256 0ce69700e301dfbbd7d27b5f9a426d4266e0a5a3b313430baa8eb1561e38b463},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.13492. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
This report presents an improved version of AlayaWorld. While the backbone architecture, chunk-wise autoregressive generation scheme, and training data remain unchanged from the previous release, we substantially revise how conditioning signals are represented and integrated into the model. The new design is guided by a simple principle: conditioning signals should match the generated content as closely as possible in both latent representation and temporal structure. To this end, we make two major changes. First, we replace the previous depth-warping-based spatial memory with a streaming 3D point-cache renderer. Second, we redesign the conditioning pipeline so that visual conditions are encoded in the same causal-VAE latent space, with temporal statistics consistent with those of the generated video. Concretely, the new version introduces six modifications: (1) replacing static-frame image conditioning with motion-aware latent conditioning; (2) causally encoding re-rendered spatial memory as a continuous sequence; (3) aligning the temporal-memory window in pixel space; (4) adopting hard memory dropout that removes memory tokens rather than zeroing them; (5) unifying the VAE encoding and decoding protocol across training and inference; and (6) removing the camera AdaLN branch, such that viewpoint control is provided entirely through the re-rendered spatial condition. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-14 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.13492 — 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 2026-10-19T04:13:06.626Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-14?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).