---
title: "VideoEditor-MoE: Scaling Complex Video Editing with Mixture of Experts"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.03043v1"
permalink: "https://feed.astralor.com/items/itm_883c73a3e07f76f5b7cebb7f"
published: "2026-11-02T12:42:51.000Z"
score: 57
channel: "tech"
tags: ["推理效率","多模态","VideoEditor-MoE","Scaling Complex Video","Editing","Mixture","Experts Complex","Meeting"]
---

# VideoEditor-MoE: Scaling Complex Video Editing with Mixture of Experts

## 速览

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 intr…

## 关键点

- 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.

## 全文

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.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_883c73a3e07f76f5b7cebb7f