HAM: A Training-Free Style Transfer Approach via Heterogeneous Attention Modulation for Diffusion Models

ArXi:2603.24043v1 Announce Type: new Diffusion models have nstrated remarkable performance in image generation, particularly within the domain of style transfer. Prevailing style transfer approaches typically leverage pre-trained diffusion models' robust feature extraction capabilities alongside external modular control pathways to explicitly impose style guidance signals. However, these methods often fail to capture complex style reference or retain the identity of user-provided content images, thus falling into the trap of style-content balance. Thus, we propose a