Open access
Sep 2026
Controllable Training-Free Diffusion Style Transfer via Latent Initialization and Attention Statistical Alignment
Recent advances in diffusion models have demonstrated remarkable generative capabilities, but their application to style transfer remains limited by inference instability and poor adaptation to diverse styles. Many methods either rely on costly fine-tuning or sacrifice flexibility and controllability at inference time....
Yan-Zhi Yuan, Zhi-Qiang Pan, Le Xia et al.
· Applied Sciences · 0 citations