Metraipolis › archive › dc_ffaa0ff4a41d

VideoEditor-MoE: Scaling Complex Video Editing with Mixture of Experts — arXiv:2611.03043

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.03043
License (as archived): CC0-1.0
Archived: 2026-11-03T11:08:36.855Z
SHA-256 of the archived text: 637589e369869c16e382bc8a9c566d3ebbae3a34c33cae7762a03ff4eee85f46
Stable URL of this page: https://metraipolis.com/documento/dc_ffaa0ff4a41d

How to cite

VideoEditor-MoE: Scaling Complex Video Editing with Mixture of Experts — arXiv:2611.03043 — arXiv (en), https://arxiv.org/abs/2611.03043 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ffaa0ff4a41d (Metraipolis archive dc_ffaa0ff4a41d, sha256 637589e369869c16e382bc8a9c566d3ebbae3a34c33cae7762a03ff4eee85f46)

BibTeX

@misc{dcffaa0ff4a41d,
  title = {VideoEditor-MoE: Scaling Complex Video Editing with Mixture of Experts — arXiv:2611.03043},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.03043},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ffaa0ff4a41d; sha256 637589e369869c16e382bc8a9c566d3ebbae3a34c33cae7762a03ff4eee85f46},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03043. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Complex video editing requires understanding complex instructions while jointly fulfilling multiple editing requests. Meeting these demands through dense model scaling substantially increases computational cost. We introduce VideoEditor-MoE, the first open-source mixture-of-experts framework for video editing. Beyond sparse MLP-MoE layers, MoE-Attention expands attention capacity through routed key--value experts while preserving global interactions among video tokens. Initialized from a 5B video editing model, VideoEditor-MoE scales to 14B and 56B parameters, with approximately 8B parameters activated per token in both variants. Comparisons with existing methods and two complementary baselines, Explicit-MoE and MLP-MoE, assess its benefits over explicit expert guidance and MLP-only expert architectures. Further analyses demonstrate that all experts remain active and reveal that learned routing responds more strongly to sample content and diffusion noise levels than to human-defined editing categories. Spatial analysis further indicates sensitivity to object-level semantic features. We also identify decoupled weight decay as a strategy for stabilizing expert scaling.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.03043 — 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 2026-11-03T10:59:46.149Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).