# ACE-LoRA: Adaptive Orthogonal Decoupling for Continual Image Editing

Source: [ACE-LoRA: Adaptive Orthogonal Decoupling for Continual Image Editing](<https://www.sota2.com/research/paper/6a06a8c0dbc836b87d82a2a7>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-16T05:00:12.406Z

Authors: Yuehao Liu, Weijia Zhang, Xuanming Shang, Zhizhou Chen, Yanhao Ge, Shanyan Guan, Chao Ma

Published: 2026-05-14T15:20:27.000Z

## Abstract

State-of-the-art diffusion models often rely on parameter-efficient fine-tuning to perform specialized image editing tasks. However, real-world applications require continual adaptation to new tasks while preserving previously learned knowledge. Despite the practical necessity, continual learning for image editing remains largely underexplored. We propose ACE-LoRA, a dynamic regularization framework for continual image editing that effectively mitigates catastrophic forgetting. ACE-LoRA leverages Adaptive Orthogonal Decoupling to identify and orthogonalize task interference, and introduces a Rank-Invariant Historical Information Compression strategy to address scalability issues in continual updates. To facilitate continual learning in image editing and provide a standardized evaluation protocol, we introduce CIE-Bench, the first comprehensive benchmark in this domain. CIE-Bench encompasses diverse and practically relevant image editing scenarios with a balanced level of difficulty to effectively expose limitations of existing models while remaining compatible with parameter-efficient fine-tuning. Extensive experiments demonstrate that our method consistently outperforms existing baselines in terms of instruction fidelity, visual realism, and robustness to forgetting, establishing a strong foundation for continual learning in image editing.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Continual Image Editing on CIE-Bench Last](<https://www.sota2.com/research/sota/continual-image-editing-on-cie-bench-last.md>): Continual Image Editing; CIE-Bench Last; ERP Score: 8.4274.

- [Continual Image Editing on CIE-Bench Avg](<https://www.sota2.com/research/sota/continual-image-editing-on-cie-bench-avg.md>): Continual Image Editing; CIE-Bench Avg; ERP Score: 8.6344.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2605.14948>)

[PDF](<https://arxiv.org/pdf/2605.14948.pdf>)
