GRNEdit：在生成式精修網路中以新二值證據視角實現高效通用影片編輯

摘要

基於指令的通用影片編輯旨在於單一直觀介面中統一多樣化的編輯操作。現有方法通常依賴資源密集型的條件機制，或採用重量級分支，或採用高成本的來源串接。是否存在一種高效方式來建模編輯意圖？為此，我們提出 GRNEdit，一個輕量級兩階段框架。GRN 透過位元組合編碼視覺語義，啟發了我們的方法。透過任務特定微調，我們進一步將此表示延伸，並將編輯語義重新表述為針對個別位元的局部保留或翻轉決策。來源資訊因而被建模為支援觀測二元狀態的逐座標證據，而 GRN 主幹網路仍負責將這些證據的全局組合解析為一致的生成語義。在第一階段，一個緊湊的編碼器將離散來源碼轉換為連續證據訊號，GRN 在二元細化過程中將其整合。受無分類器引導中空提示訓練的啟發，我們進一步賦予空條件一種編輯特定的含義：空指令表示不編輯，並透過來源重建進行監督。這條恆等路徑不僅在第一階段隱式地增強了證據利用與內容保留，還產生了與編輯狀態處於同一表示空間中的來源保留狀態。因此，第二階段可直接比較每個編輯狀態與其對應的來源保留狀態，並利用兩者之間的差異來修訂尚未解決的目標位元決策。僅使用 0.6M 對訓練資料且條件參數少於 3% 的情況下，GRNEdit-2B 與 GRNEdit-8B 在 OpenVE-Bench 上分別達到 4.03 與 4.18 的分數。其中，2B 模型優於多個 14B 開源編輯器，而 8B 模型則與領先的開源編輯器表現相當。

English

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.