Censored Sampling of Diffusion Models Using 3 Minutes of Human Feedback
Abstract
Diffusion models have recently shown remarkable success in high-quality image generation. Sometimes, however, a pre-trained diffusion model exhibits partial misalignment in the sense that the model can generate good images, but it sometimes outputs undesirable images. If so, we simply need to prevent the generation of the bad images, and we call this task censoring. In this work, we present censored generation with a pre-trained diffusion model using a reward model trained on minimal human feedback. We show that censoring can be accomplished with extreme human feedback efficiency and that labels generated with a mere few minutes of human feedback are sufficient.
Cite
Text
Yoon et al. "Censored Sampling of Diffusion Models Using 3 Minutes of Human Feedback." Neural Information Processing Systems, 2023.Markdown
[Yoon et al. "Censored Sampling of Diffusion Models Using 3 Minutes of Human Feedback." Neural Information Processing Systems, 2023.](https://mlanthology.org/neurips/2023/yoon2023neurips-censored/)BibTeX
@inproceedings{yoon2023neurips-censored,
title = {{Censored Sampling of Diffusion Models Using 3 Minutes of Human Feedback}},
author = {Yoon, TaeHo and Myoung, Kibeom and Lee, Keon and Cho, Jaewoong and No, Albert and Ryu, Ernest},
booktitle = {Neural Information Processing Systems},
year = {2023},
url = {https://mlanthology.org/neurips/2023/yoon2023neurips-censored/}
}