Beyond Generation: Unlocking Universal Editing via Self-Supervised Fine-Tuning

TL;DR

This paper introduces UES, a self-supervised fine-tuning method that transforms video generation models into versatile editing systems, reducing training costs and enabling universal editing capabilities across diverse tasks.

Contribution

The paper presents a lightweight self-supervised fine-tuning approach that unifies video generation and editing, significantly reducing training complexity and enabling broad applicability.

Findings

Enables models to perform universal editing without additional supervision.

Reduces tunable parameters by over 92%.

Achieves state-of-the-art editing performance across diverse tasks.

Abstract

Recent advances in video generation have outpaced progress in video editing, which remains constrained by several limiting factors, namely: (a) the task's dependency on supervision severely limits generality, (b) an unnecessary artificial separation between the generation and editing task, and (c) the high computational costs of training a video model. In this work, we propose UES (Unlocking Universal Editing via Self-Supervision), a lightweight self-supervised fine-tuning strategy that transforms generation models into unified generation-editing systems through self-supervised semantic alignment. Our approach establishes a dual-conditioning mechanism where original video-text pairs jointly provide visual and textual semantics, enabling structured learning of intrinsic spatiotemporal correspondences. Key advantages include: (i) Universality through supervision-free adaptation to diverse editing tasks, (ii) Unification of generation and editing applicable to most text(+image)-to-video model, and (iii) Efficiency via lightweight fine-tune that reduces tunable parameters by 92.67%. To enable systematic evaluation, we introduce OmniBench-99, a comprehensive benchmark spanning 99 videos across humans/animals, environments, and objects, comprising 4 editing types and 8 scenarios. Extensive experiments show UES enables models without inherent editing capability to perform powerful and universal editing while preserving or even enhancing their original generation performance.