Dynamic Negative Guidance of Diffusion Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Koulischer, Felix, Deleu, Johannes, Raya, Gabriel, Demeester, Thomas, Ambrogioni, Luca
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915335642284032
author Koulischer, Felix
Deleu, Johannes
Raya, Gabriel
Demeester, Thomas
Ambrogioni, Luca
author_facet Koulischer, Felix
Deleu, Johannes
Raya, Gabriel
Demeester, Thomas
Ambrogioni, Luca
contents Negative Prompting (NP) is widely utilized in diffusion models, particularly in text-to-image applications, to prevent the generation of undesired features. In this paper, we show that conventional NP is limited by the assumption of a constant guidance scale, which may lead to highly suboptimal results, or even complete failure, due to the non-stationarity and state-dependence of the reverse process. Based on this analysis, we derive a principled technique called Dynamic Negative Guidance, which relies on a near-optimal time and state dependent modulation of the guidance without requiring additional training. Unlike NP, negative guidance requires estimating the posterior class probability during the denoising process, which is achieved with limited additional computational overhead by tracking the discrete Markov Chain during the generative process. We evaluate the performance of DNG class-removal on MNIST and CIFAR10, where we show that DNG leads to higher safety, preservation of class balance and image quality when compared with baseline methods. Furthermore, we show that it is possible to use DNG with Stable Diffusion to obtain more accurate and less invasive guidance than NP.
format Preprint
id arxiv_https___arxiv_org_abs_2410_14398
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Dynamic Negative Guidance of Diffusion Models
Koulischer, Felix
Deleu, Johannes
Raya, Gabriel
Demeester, Thomas
Ambrogioni, Luca
Computer Vision and Pattern Recognition
Negative Prompting (NP) is widely utilized in diffusion models, particularly in text-to-image applications, to prevent the generation of undesired features. In this paper, we show that conventional NP is limited by the assumption of a constant guidance scale, which may lead to highly suboptimal results, or even complete failure, due to the non-stationarity and state-dependence of the reverse process. Based on this analysis, we derive a principled technique called Dynamic Negative Guidance, which relies on a near-optimal time and state dependent modulation of the guidance without requiring additional training. Unlike NP, negative guidance requires estimating the posterior class probability during the denoising process, which is achieved with limited additional computational overhead by tracking the discrete Markov Chain during the generative process. We evaluate the performance of DNG class-removal on MNIST and CIFAR10, where we show that DNG leads to higher safety, preservation of class balance and image quality when compared with baseline methods. Furthermore, we show that it is possible to use DNG with Stable Diffusion to obtain more accurate and less invasive guidance than NP.
title Dynamic Negative Guidance of Diffusion Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2410.14398