Metraipolis › archive › dc_ea598b1be86f
WorldAttention: An Efficient Attention Architecture for Interactive Video World Models — arXiv:2609.34606
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.34606
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T23:52:18.096Z
- SHA-256 of the archived text
4c92f5dca025187413eeff93ede8437ae508f16eea9ad1767fa2b178e65870d3- Stable URL of this page
https://metraipolis.com/documento/dc_ea598b1be86f
How to cite
WorldAttention: An Efficient Attention Architecture for Interactive Video World Models — arXiv:2609.34606 — arXiv (en), https://arxiv.org/abs/2609.34606 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ea598b1be86f (Metraipolis archive dc_ea598b1be86f, sha256 4c92f5dca025187413eeff93ede8437ae508f16eea9ad1767fa2b178e65870d3)
BibTeX
@misc{dcea598b1be86f,
title = {WorldAttention: An Efficient Attention Architecture for Interactive Video World Models — arXiv:2609.34606},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.34606},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ea598b1be86f; sha256 4c92f5dca025187413eeff93ede8437ae508f16eea9ad1767fa2b178e65870d3},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.34606. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Leveraging the paradigm of autoregressive diffusion, text-conditioned interactive video world models aim to simulate temporally coherent environments guided by textual instructions. While enabling low-latency, long-duration generation is pivotal for embodied AI and simulation-based planning, current frameworks primarily rely on sliding-window mechanisms to bound computational complexity. However, this approach inherently sacrifices historical context, undermining the long-range interactive capabilities. Conversely, maintaining a full-history cache remains computationally prohibitive and memory-intensive: the quadratic complexity of attention leads to excessive computational overhead, while the linear growth of the KV cache inevitably leads to GPU memory saturation. To overcome these limitations, we propose WorldAttention, a system-oriented attention architecture that achieves high efficiency through the co-design of specialized attention kernels and hierarchical KV cache management. First, we introduce Hybrid Sparse Attention (HSA), which integrates linear global attention supplemented with head-adaptive sparse attention. Additionally, we design a Hierarchical KV Cache (HKV) that organizes historical KV pairs into semantically indexed pages across multi-tier memory, enabling fine-grained retrieval and controlled GPU residency. These two designs are supported by tailored kernels to effectively translate their theoretical efficiency into real-world performance. Extensive experiments on VBench-Long and InterVBench demonstrate that WorldAttention consistently surpasses prior state-of-the-art methods, achieving subject consistency scores of 0.9472 on VBench-Long and 0.9668 on InterVBench, respectively. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.34606 — 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 2026-10-07T23:29:27.407Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).