StyleQoRA: Quality-Aware Low-Rank Adaptation for Few-Shot Multi-Style Editing

TL;DR

StyleQoRA introduces a novel, quality-aware low-rank adaptation framework for effective few-shot multi-style image editing, leveraging automatic rank determination and a mixture-of-experts approach to outperform existing methods.

Contribution

The paper proposes StyleQoRA, a new framework that automatically determines layer-wise ranks and uses hybrid routing to improve few-shot multi-style image editing.

Findings

Outperforms state-of-the-art methods in multi-style editing

Uses fewer LoRA parameters for comparable or better results

Effectively balances style-specific and shared knowledge

Abstract

In recent years, image editing has garnered growing attention. However, general image editing models often fail to produce satisfactory results when confronted with new styles. The challenge lies in how to effectively fine-tune general image editing models to new styles using only a limited amount of paired data and a minimum number of parameters. To address this issue, this paper proposes a novel few-shot multi-style editing framework. For this task, we construct a benchmark dataset that encompasses five distinct styles. Correspondingly, we propose Quality-Aware Low-Rank Adaptation for few-shot multi-style editing (StyleQoRA). Our StyleQoRA can automatically determine the optimal rank for each layer through a novel approach that estimates the importance score of each single-rank component using an image quality metric. To balance specialization and knowledge sharing, we design a Mixture-of-Experts (MoE) LoRA with hybrid routing in our StyleQoRA, consisting of style-specific routing to prevent cross-style confusion and style-shared routing to capture common transformation patterns. Additionally, we explore the optimal location to insert LoRA within the Diffusion in Transformer (DiT) model and integrate adversarial learning and flow matching to guide the diffusion training process. Experimental results demonstrate that our proposed method outperforms existing state-of-the-art approaches with significantly fewer LoRA parameters. Our code and dataset are available at https://github.com/cao-cong/FSMSE.