Metraipolis › archive › dc_a07cb804f481

ReelCache: An Efficient KV Cache Compression and Management System for Multi-Round Video Generation — arXiv:2611.09973

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.09973
License (as archived): CC0-1.0
Archived: 2026-11-10T18:45:54.112Z
SHA-256 of the archived text: 13e3067028aad232e16928160a122beb29790cd94eb51cf9c9ba35019fb689e5
Stable URL of this page: https://metraipolis.com/documento/dc_a07cb804f481

How to cite

ReelCache: An Efficient KV Cache Compression and Management System for Multi-Round Video Generation — arXiv:2611.09973 — arXiv (en), https://arxiv.org/abs/2611.09973 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a07cb804f481 (Metraipolis archive dc_a07cb804f481, sha256 13e3067028aad232e16928160a122beb29790cd94eb51cf9c9ba35019fb689e5)

BibTeX

@misc{dca07cb804f481,
  title = {ReelCache: An Efficient KV Cache Compression and Management System for Multi-Round Video Generation — arXiv:2611.09973},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.09973},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a07cb804f481; sha256 13e3067028aad232e16928160a122beb29790cd94eb51cf9c9ba35019fb689e5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.09973. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Autoregressive video diffusion models produce large KV caches during multi-round video generation. Keeping all these KV chunks would consume substantial memory, while rebuilding discarded caches would significantly delay video chunk generation. This paper presents ReelCache, a KV cache compression system for efficient multi-round, long video generation. ReelCache combines an intra-chunk compressor with an inter-chunk cache manager. The compressor applies low-rank decomposition across channels and video coding to the resulting coefficients, predicting the decomposition rank per transformer layer online to meet retained-energy targets. The cache manager shares basis directions across consecutive chunks and organizes chunk states in a tree, allowing generation branches to share historical state while supporting restoration for resume, rewind, and fork. Experiments across five video generation models show that ReelCache reduces KV storage by 14.46-21.13x relative to baselines and accelerates next-chunk generation by 1.25-3.36x, with less than 1% drop in six representative VBench metrics. GitHub: this https URL

Tema degli esploratori: calcolo e infrastrutture (elenco arXiv cs.DC, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Distributed, Parallel, and Cluster Computing (cs.DC).
Pagina del lavoro: https://arxiv.org/abs/2611.09973 — 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 2026-11-10T18:38:07.449Z dalla pagina https://arxiv.org/list/cs.DC/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).