Abstract
Image editing should change only the attributes specified by an instruction while preserving everything else, yet current methods often make unintended changes. We treat this minimal-change principle as an optimization objective for instruction-based editing. Latent L1 regularization is a poor proxy for output locality in modern nonlinear generators and often requires supervision unavailable at scale. We instead optimize edit outcomes with reinforcement learning. An agentic vision-language reward model audits each source image, instruction, and edited image for two failure types: unimplemented requested changes and unintended changes. A group-level rubric merges and verifies these issues to provide consistent rewards across candidate edits without per-instruction human annotations. On FLUX.1 Kontext-dev, ARRO raises average EditScore from 5.21 to 5.88 across MinEval, MagicBrush, AnyBench, and Emu-Edit. On 600 evaluation examples, it reduces off-target pixel change by 8.4% relative to the base editor. Reward and SFT controls, blinded human evaluations, and transfer to OmniGen2 provide complementary evidence. Code: https://github.com/Showwwwwwwww/ARRO
Keywords
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Chen, S., Huang, F., Yao, Y., Gong, M., & Liu, T. (2026). Scalable Minimal-Change Learning for Controllable Image Editing. https://omanscience.com/en/articles/scalable-minimal-change-learning-for-controllable-image-editing
MLA 9
Chen, Shuo, et al. "Scalable Minimal-Change Learning for Controllable Image Editing." https://omanscience.com/en/articles/scalable-minimal-change-learning-for-controllable-image-editing.
Chicago (author–date)
Chen, Shuo, Fengming Huang, Yu Yao, Mingming Gong, and Tongliang Liu. 2026. "Scalable Minimal-Change Learning for Controllable Image Editing." https://omanscience.com/en/articles/scalable-minimal-change-learning-for-controllable-image-editing.
Harvard
Chen, S., Huang, F., Yao, Y., Gong, M. and Liu, T. (2026) 'Scalable Minimal-Change Learning for Controllable Image Editing', Available at: https://omanscience.com/en/articles/scalable-minimal-change-learning-for-controllable-image-editing.
Vancouver
Chen S, Huang F, Yao Y, Gong M, Liu T. Scalable Minimal-Change Learning for Controllable Image Editing. https://omanscience.com/en/articles/scalable-minimal-change-learning-for-controllable-image-editing
IEEE
S. Chen, F. Huang, Y. Yao, M. Gong, and T. Liu, "Scalable Minimal-Change Learning for Controllable Image Editing," https://omanscience.com/en/articles/scalable-minimal-change-learning-for-controllable-image-editing.