paper-with-me

홈 › Papers

FREE-Switch: Frequency-based Dynamic LoRA Switch for Style Transfer

2026-04-11 · Shenghe Zheng, Minyu Zhang, Tianhao Liu, Hongzhi Wang arxiv

With the growing availability of open-sourced adapters trained on the same diffusion backbone for diverse scenes and objects, combining these pretrained weights enables low-cost customized generation. However, most existing model merging methods are designed for classification or text generation, and when applied to image generation, they suffer from content drift due to error accumulation across multiple diffusion steps. For image-oriented methods, training-based approaches are computationally expensive and unsuitable for edge deployment, while training-free ones use uniform fusion strategies that ignore inter-adapter differences, leading to detail degradation. We find that since different adapters are specialized for generating different types of content, the contribution of each diffusion step carries different significance for each adapter. Accordingly, we propose a frequency-domain importance-driven dynamic LoRA switch method. Furthermore, we observe that maintaining semantic consistency across adapters effectively mitigates detail loss; thus, we design an automatic Generation Alignment mechanism to align generation intents at the semantic level. Experiments demonstrate that our FREE-Switch (Frequency-based Efficient and Dynamic LoRA Switch) framework efficiently combines adapters for different objects and styles, substantially reducing the training cost of high-quality customized generation.

📄 PDF Abstract BibTeX arXiv:2604.10023

Code (0)

등록된 구현이 없습니다.

Tasks

Image GenerationText GenerationStyle Transfer

Similar Papers 제목 키워드 기반

SwitchLoRA: Switched Low-Rank Adaptation Can Learn Full-Rank Information

2024-06-03 · Kaiye Zhou, Shucheng Wang, Jun Xu

In the training of large language models, parameter-efficient techniques such as LoRA optimize memory usage and reduce communication overhead and memory usage during the fine-tuning phase. However, applying such techniqu…

Language ModellingLarge Language Modelparameter-efficient fine-tuning

LoRA-Switch: Boosting the Efficiency of Dynamic LLM Adapters via System-Algorithm Co-design

2024-05-28 · Rui Kong, Qiyang Li, Xinyu Fang, Qingtian Feng 외

Recent literature has found that an effective method to customize or further improve large language models (LLMs) is to add dynamic adapters, such as low-rank adapters (LoRA) with Mixture-of-Experts (MoE) structures. Tho…

Mixture-of-Experts

Learning Invariant Stabilizing Controllers for Frequency Regulation under Variable Inertia

2022-03-09 · Priyank Srivastava, Patricia Hidalgo-Gonzalez, Jorge Cortes

Declines in cost and concerns about the environmental impact of traditional generation have boosted the penetration of renewables and non-conventional distributed energy resources into the power grid. The intermittent av…

Sample-Efficient Reinforcement Learning with loglog(T) Switching Cost

2022-02-13 · Dan Qiao, Ming Yin, Ming Min, Yu-Xiang Wang

We study the problem of reinforcement learning (RL) with low (policy) switching cost - a problem well-motivated by real-life RL applications in which deployments of new policies are costly and the number of policy update…

reinforcement-learningReinforcement LearningReinforcement Learning (RL)

Provably Efficient Q-Learning with Low Switching Cost

2019-05-30 · NeurIPS 2019 12 · Yu Bai, Tengyang Xie, Nan Jiang, Yu-Xiang Wang

We take initial steps in studying PAC-MDP algorithms with limited adaptivity, that is, algorithms that change its exploration policy as infrequently as possible during regret minimization. This is motivated by the diffic…

Q-Learning