Metraipolis › archive › dc_fa92cc8d002b

Multi-Grid Post-Training for Long-Form Multi-Shot Video Generation — arXiv:2609.06373

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.06373
License (as archived): CC0-1.0
Archived: 2026-10-16T15:06:59.974Z
SHA-256 of the archived text: 011f94468a4989069d46dbe6bd0a7158ad2213e0b8d2537b1ce2f57685eddf13
Stable URL of this page: https://metraipolis.com/documento/dc_fa92cc8d002b

How to cite

Multi-Grid Post-Training for Long-Form Multi-Shot Video Generation — arXiv:2609.06373 — arXiv (en), https://arxiv.org/abs/2609.06373 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fa92cc8d002b (Metraipolis archive dc_fa92cc8d002b, sha256 011f94468a4989069d46dbe6bd0a7158ad2213e0b8d2537b1ce2f57685eddf13)

BibTeX

@misc{dcfa92cc8d002b,
  title = {Multi-Grid Post-Training for Long-Form Multi-Shot Video Generation — arXiv:2609.06373},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.06373},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fa92cc8d002b; sha256 011f94468a4989069d46dbe6bd0a7158ad2213e0b8d2537b1ce2f57685eddf13},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.06373. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Generating long-form multi-shot videos requires coherent within-shot motion and visually consistent narratives across shots. Existing video generators favor continuous motion and struggle to present complete shot sets when an entire narrative is packed along one temporal axis. We propose MovieGrid, a Multi-Grid Post-Training paradigm that decomposes a long video into shorter, temporally ordered chunks and arranges them on a spatial grid for joint modeling. This design reduces the number of shots handled by each temporal axis while enabling global information exchange across chunks. We construct the Multi-Grid Long Video (MGLV) dataset from 1,000 long-form videos using source video collection, hierarchical segmentation, grid video construction, and character-aware story annotation, producing 54K grid videos paired with story prompts. Our Noise-Free Random-Grid Training retains a random subset of chunks as clean visual context for denoising the remaining chunks. Grid Embedding encodes grid structure, character-aware Story Prompts link recurring entities, and Grid Boundary Loss stabilizes layouts. Under the same token budget, MovieGrid generates 6.05 times more shots than Temporal Packing in a 1,616-frame video. On a benchmark spanning five real-world categories, it achieves state-of-the-art intra-shot consistency (0.9131 versus 0.8086 for HoloCine) and inter-shot consistency (0.5914 versus 0.5384 for StoryMem). MovieGrid can further scale video length with minimal compromise through single or multiple generations.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.06373 — 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 2026-10-16T14:49:57.080Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).