Edit-GRPO: A Locality-Preserving Policy Optimization Framework for Image Editing

TL;DR

Edit-GRPO is a novel framework that enhances image editing by explicitly preserving spatial locality through region-specific policy optimization, reducing artifacts and maintaining visual coherence.

Contribution

It introduces a locality-preserving policy optimization framework that decouples editing and preservation objectives for improved localized image editing.

Findings

Significantly improves locality preservation in image editing.

Reduces artifacts such as context distortion and boundary inconsistency.

Maintains strong editing performance across diverse scenarios.

Abstract

A fundamental challenge in image editing lies in preserving spatial locality: edits should improve targeted content without inadvertently altering surrounding regions. However, most optimization-based editing approaches treat images as holistic entities, causing global policy updates that undermine locality and introduce undesired context changes. We observe that this issue stems from a mismatch between localized editing intent and globally applied optimization signals. Motivated by this insight, we propose Edit-GRPO, preserving Locality while optimizing image editing, a locality-preserving policy optimization framework that explicitly decouples editing and preservation objectives. By assigning region-specific optimization signals to edit and non-edit areas, Edit-GRPO aligns policy updates with the spatial structure of editing tasks, enabling localized improvements while maintaining global visual coherence. This design effectively suppresses common artifacts such as context distortion and boundary inconsistency. Extensive experiments across diverse image editing scenarios demonstrate that Edit-GRPO significantly improves locality preservation while maintaining strong editing performance compared to existing optimization-based methods, validating the generality and effectiveness of the proposed framework.