Positive-Unlabeled Diffusion Models for Preventing Sensitive Data Generation

Hiroshi Takahashi, Tomoharu Iwata, Atsutoshi Kumagai, Yuuki Yamanaka, Tomoya Yamashita

Keywords: Diffusion Model, Positive-Unlabeled Learning

TL;DR: We prevent diffusion models from generating sensitive data by using positive-unlabeled learning.

Abstract:

Diffusion models are powerful generative models but often generate sensitive data that are unwanted by users,
mainly because the unlabeled training data frequently contain such sensitive data.
Since labeling all sensitive data in the large-scale unlabeled training data is impractical,
we address this problem by using a small amount of labeled sensitive data.
In this paper,
we propose positive-unlabeled diffusion models,
which prevent the generation of sensitive data using unlabeled and sensitive data.
Our approach can approximate the evidence lower bound (ELBO) for normal (negative) data using only unlabeled and sensitive (positive) data.
Therefore, even without labeled normal data,
we can maximize the ELBO for normal data and minimize it for labeled sensitive data,
ensuring the generation of only normal data.
Through experiments across various datasets and settings,
we demonstrated that our approach can prevent the generation of sensitive images without compromising image quality.

Primary Area: generative models

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 7852