Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

FeRA: Frequency-Energy Constrained Routing for Effective Diffusion Adaptation Fine-Tuning

About

Diffusion models have achieved remarkable success in generative modeling, yet how to effectively adapt large pretrained models to new tasks remains challenging. We revisit the reconstruction behavior of diffusion models during denoising to unveil the underlying frequency energy mechanism governing this process. Building upon this observation, we propose FeRA, a frequency driven fine tuning framework that aligns parameter updates with the intrinsic frequency energy progression of diffusion. FeRA establishes a comprehensive frequency energy framework for effective diffusion adaptation fine tuning, comprising three synergistic components: (i) a compact frequency energy indicator that characterizes the latent bandwise energy distribution, (ii) a soft frequency router that adaptively fuses multiple frequency specific adapter experts, and (iii) a frequency energy consistency regularization that stabilizes diffusion optimization and ensures coherent adaptation across bands. Routing operates in both training and inference, with inference time routing dynamically determined by the latent frequency energy. It integrates seamlessly with adapter based tuning schemes and generalizes well across diffusion backbones and resolutions. By aligning adaptation with the frequency energy mechanism, FeRA provides a simple, stable, and compatible paradigm for effective and robust diffusion model adaptation.

Bo Yin, Xiaobin Hu, Xingyu Zhou, Yu He, Peng-Tao Jiang, Yue Liao, Junwei Zhu, Jiangning Zhang, Ying Tai, Shuicheng Yan• 2025

Related benchmarks

TaskDatasetResultRank
Diffusion AdaptationBarbie
CLIP Score37.23
48
Diffusion AdaptationCyberpunk
CLIP Score34.9
48
Diffusion AdaptationExpedition
CLIP Score33.98
48
Diffusion AdaptationHornify
CLIP Score33.95
48
Diffusion AdaptationElementFire
CLIP Score33.82
48
Personalized Image GenerationDreamBooth--
45
DreamBooth-style Image CustomizationBarbie
CLIP Score37.86
32
DreamBooth-style Image CustomizationCyberpunk
CLIP Score34.73
32
DreamBooth-style Image CustomizationExpedition
CLIP Score33.22
32
DreamBooth-style Image CustomizationHornify
CLIP Score33.63
32
Showing 10 of 17 rows

Other info

Follow for update