English

Towards Controllable Image Generation through Representation-Conditioned Diffusion Models

Computer Vision and Pattern Recognition 2026-05-27 v1 Machine Learning

Abstract

Diffusion models have emerged as powerful tools for high-quality image generation and editing, but guiding these models to produce specific outputs remains a challenge. Conventional approaches rely on conditioning mechanisms, such as text prompts or semantic maps, which require extensively annotated datasets. In this preliminary work, we explore diffusion models conditioned on representations from a pre-trained self-supervised model. The self-conditioning mechanism not only improves the quality of unconditional image generation, but also provides a representation space that can be used to control the generation. We explore this conditioning space by identifying directions of variations, and demonstrate promising properties in terms of smoothness and disentanglement.

Keywords

Cite

@article{arxiv.2605.27343,
  title  = {Towards Controllable Image Generation through Representation-Conditioned Diffusion Models},
  author = {Nithesh Chandher Karthikeyan and Jonas Unger and Gabriel Eilertsen},
  journal= {arXiv preprint arXiv:2605.27343},
  year   = {2026}
}