GRNEdit: Efficient General Video Editing from a New Binary-Evidence Perspective in Generative Refinement Networks

📄 arXiv: 2608.16328v1 📥 PDF

作者: Feng Xie, Jiagao Hu, Fuhao Li, Zepeng Wang, Yuxuan Chen, Dahua Gao, Fei Wang, Daiguo Zhou

分类: cs.CV

发布日期: 2026-08-17

💡 一句话要点

提出GRNEdit以解决高效视频编辑问题

🎯 匹配领域: 支柱四：生成式动作 (Generative Motion)

关键词: 视频编辑 生成网络 轻量级框架 比特编码 内容保留 编辑语义 高效处理

📋 核心要点

  1. 现有视频编辑方法往往依赖于复杂的条件设置，导致资源消耗高，效率低下。
  2. GRNEdit通过轻量级的两阶段框架，采用比特组合编码视觉语义，简化了编辑意图建模。
  3. 在仅使用少量训练数据的情况下，GRNEdit在多个基准测试中表现优异，超越了多个开源编辑器。

📝 摘要（中文）

基于指令的一般视频编辑旨在通过单一直观界面统一多种编辑操作。现有方法通常依赖于资源密集型的条件设置，使用重型分支或昂贵的源连接。为此，本文提出GRNEdit，一个轻量级的两阶段框架。通过任务特定的微调，本文将编辑语义重新构建为对单个比特的局部保留或翻转决策。GRNEdit在仅使用0.6M对训练数据和不到3%的条件参数的情况下，取得了在OpenVE-Bench上4.03和4.18的得分，展示了其在性能上的优势。

🔬 方法详解

问题定义：现有的视频编辑方法通常需要复杂的条件设置，导致资源消耗高且效率低下，难以实现快速和高效的编辑操作。

核心思路：GRNEdit通过引入轻量级的两阶段框架，利用比特组合来编码视觉语义，从而简化编辑意图的建模过程。该方法通过局部的保留或翻转决策来处理编辑语义，提升了编辑效率。

技术框架：GRNEdit的整体架构分为两个阶段：第一阶段是一个紧凑的编码器，将离散源代码转换为连续的证据信号；第二阶段则比较编辑状态与源保留状态的差异，修正目标比特决策。

关键创新：GRNEdit的主要创新在于通过比特的局部决策来处理编辑语义，并引入了空条件的编辑特定意义，增强了证据利用和内容保留。

关键设计：在训练过程中，GRNEdit使用了少量的训练对（0.6M），并且条件参数少于3%。通过源重建监督空指令，强化了第一阶段的证据利用和内容保留。

🖼️ 关键图片

fig_0
fig_1
fig_2

📊 实验亮点

GRNEdit在OpenVE-Bench上取得了4.03和4.18的得分，2B模型超越了多个14B的开源编辑器，而8B模型与领先的开源编辑器表现相当，展示了其在性能上的显著提升。

🎯 应用场景

GRNEdit在视频编辑领域具有广泛的应用潜力，能够为内容创作者提供高效、直观的编辑工具。其轻量级设计使其适用于实时视频处理和在线编辑平台，未来可能在社交媒体、影视制作等多个领域产生重要影响。

📄 摘要（原文）

Instruction-based general video editing seeks to unify diverse editing operations within a single, intuitive interface. Existing approaches often rely on resource-intensive conditioning, using either heavyweight branches or costly source concatenation. Is there any efficient way to model editing intent? Thus, we introduce GRNEdit, a lightweight two-stage framework. GRN inspires our approach by encoding visual semantics through combinations of bits. Through task-specific fine-tuning, we take this representation further and recast editing semantics as local retain-or-flip decisions over individual bits. Source information is consequently modeled as coordinate-wise evidence supporting the observed binary states, while the GRN backbone remains responsible for resolving their global composition into coherent generative semantics. In Stage I, a compact encoder translates discrete source codes into continuous evidence signals, which GRN assimilates throughout binary refinement. Inspired by null-prompt training for classifier-free guidance, we further assign the null condition an editing-specific meaning: an empty instruction denotes no edit and is supervised through source reconstruction. This identity pathway not only implicitly strengthens evidence utilization and content preservation in Stage I, but also produces a source-preserving state in the same representation space as the edited state. Stage II can therefore directly compare each edited state with its source-preserving counterpart and use their discrepancy to revise unresolved target-bit decisions. Trained on only 0.6M pairs with less than 3\% conditioning parameters, GRNEdit-2B and GRNEdit-8B achieve scores of 4.03 and 4.18 on OpenVE-Bench. The 2B model outperforms multiple 14B open-source editors, while the 8B model performs on par with leading open-source editors.