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

  • 类型:arxiv
  • 标识:2608.16328
  • 链接:https://arxiv.org/abs/2608.16328
  • 主分类:multimodal
  • 形态:method
  • 被引:0
  • 被引来源:Semantic Scholar
  • S2被引:0
  • 影响力被引:0
  • TLDR:GRNEdit, a lightweight two-stage framework for instruction-based general video editing that outperforms multiple 14B open-source editors, while the 8B model performs on par with leading open-source editors.
  • 待LLM分类:否
  • 标题中文:GRNEdit:基于生成式 Refinement 网络的新二元证据视角下的高效通用视频编辑
  • TLDR中文:GRNEdit 是一个轻量级的两阶段指令驱动通用视频编辑框架,性能优于多个 14B 开源编辑器,同时其 8B 模型与领先的开源编辑器表现相当。
  • 来源文件
  • /inbox/tom/_candidates/2026-08-18-agent-rag-longcontext-candidates.json
  • [S2 enrich]