GRNEdit: Efficient General Video Editing from a New Binary-Evidence Perspective in Generative Refinement Networks
2026-08-17 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors propose GRNEdit, a two-stage video editing framework that uses a lightweight method to represent and edit video content by treating editing decisions as flipping bits in a binary code. Their approach efficiently models the editing intent without heavy computational costs by encoding source video information as evidence signals and refining edits through comparisons with the original content. They also introduce a way to handle 'no edit' instructions to help preserve the original video details. Tested on a standard benchmark, their smaller models outperform or match larger existing editors despite using fewer parameters and less training data.
Instruction-based video editingGRN (Gated Residue Network)Binary representationTwo-stage frameworkClassifier-free guidanceSource reconstructionOpenVE-BenchVideo editing intentLightweight encoderContent preservation
Authors
Feng Xie, Jiagao Hu, Fuhao Li, Zepeng Wang, Yuxuan Chen, Dahua Gao, Fei Wang, Daiguo Zhou
Abstract
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.