Metraipolis › archive › dc_e659ba2e128e
What Helps Denoising Can Hurt Memory: Rethinking History in Long Video Generation — arXiv:2611.00937
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.00937
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T18:53:19.768Z
- SHA-256 of the archived text
00fc8f297c821760bfa000a2cdcc8a5575de5e63e902522957bc1682951bcbc5- Stable URL of this page
https://metraipolis.com/documento/dc_e659ba2e128e
How to cite
What Helps Denoising Can Hurt Memory: Rethinking History in Long Video Generation — arXiv:2611.00937 — arXiv (en), https://arxiv.org/abs/2611.00937 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e659ba2e128e (Metraipolis archive dc_e659ba2e128e, sha256 00fc8f297c821760bfa000a2cdcc8a5575de5e63e902522957bc1682951bcbc5)
BibTeX
@misc{dce659ba2e128e,
title = {What Helps Denoising Can Hurt Memory: Rethinking History in Long Video Generation — arXiv:2611.00937},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.00937},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e659ba2e128e; sha256 00fc8f297c821760bfa000a2cdcc8a5575de5e63e902522957bc1682951bcbc5},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.00937. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Autoregressive video generators produce long videos block by block, using historical context to maintain temporal consistency. After denoising, a committing pass encodes each clean block into keys and values (KV) that later blocks read, and the same clean block can be encoded into different KV depending on the history it reads. We therefore study historical context not only as a condition for denoising but also as an input to committing. Older historical context that supports denoising can harm subsequent generation when read during committing. At fixed context size and temporal positions, older historical context lowers mean clip-to-clip Subject Consistency. Across 5 generator configurations supporting longer contexts, consistency rises, plateaus, and then declines as committing context expands. Motivated by these findings, we propose Commit Forcing, a training-free rule that organizes committing into a chain. During committing, each block reads only the preceding block's stored KV, so older historical context reaches new entries only through the chain. We apply this rule to 6 pretrained generators. On 60-second VBench, Commit Forcing improves Quality and Total on all 6 generators, by up to 1.27 Quality points and 1.10 Total points. Long-range consistency gains extend to 240-second videos. In a blinded comparison on 120-second LongLive videos, Commit Forcing achieves win rates of 83.5% for subject and scene consistency and 82.9% for overall preference against the base generator, excluding ties. Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2611.00937 — 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 2026-11-03T18:16:52.910Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).