Salvato in:
Dettagli Bibliografici
Autori principali: Pavasovic, Krunoslav Lehman, Verbeek, Jakob, Biroli, Giulio, Mezard, Marc
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2502.07849
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909620208926720
author Pavasovic, Krunoslav Lehman
Verbeek, Jakob
Biroli, Giulio
Mezard, Marc
author_facet Pavasovic, Krunoslav Lehman
Verbeek, Jakob
Biroli, Giulio
Mezard, Marc
contents Classifier-Free Guidance (CFG) is a widely adopted technique in diffusion and flow-based generative models, enabling high-quality conditional generation. A key theoretical challenge is characterizing the distribution induced by CFG, particularly in high-dimensional settings relevant to real-world data. Previous works have shown that CFG modifies the target distribution, steering it towards a distribution sharper than the target one, more shifted towards the boundary of the class. In this work, we provide a high-dimensional analysis of CFG, showing that these distortions vanish as the data dimension grows. We present a blessing-of-dimensionality result demonstrating that in sufficiently high and infinite dimensions, CFG accurately reproduces the target distribution. Using our high-dimensional theory, we show that there is a large family of guidances enjoying this property, in particular non-linear CFG generalizations. We study a simple non-linear power-law version, for which we demonstrate improved robustness, sample fidelity and diversity. Our findings are validated with experiments on class-conditional and text-to-image generation using state-of-the-art diffusion and flow-matching models.
format Preprint
id arxiv_https___arxiv_org_abs_2502_07849
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Classifier-Free Guidance: From High-Dimensional Analysis to Generalized Guidance Forms
Pavasovic, Krunoslav Lehman
Verbeek, Jakob
Biroli, Giulio
Mezard, Marc
Machine Learning
Artificial Intelligence
Classifier-Free Guidance (CFG) is a widely adopted technique in diffusion and flow-based generative models, enabling high-quality conditional generation. A key theoretical challenge is characterizing the distribution induced by CFG, particularly in high-dimensional settings relevant to real-world data. Previous works have shown that CFG modifies the target distribution, steering it towards a distribution sharper than the target one, more shifted towards the boundary of the class. In this work, we provide a high-dimensional analysis of CFG, showing that these distortions vanish as the data dimension grows. We present a blessing-of-dimensionality result demonstrating that in sufficiently high and infinite dimensions, CFG accurately reproduces the target distribution. Using our high-dimensional theory, we show that there is a large family of guidances enjoying this property, in particular non-linear CFG generalizations. We study a simple non-linear power-law version, for which we demonstrate improved robustness, sample fidelity and diversity. Our findings are validated with experiments on class-conditional and text-to-image generation using state-of-the-art diffusion and flow-matching models.
title Classifier-Free Guidance: From High-Dimensional Analysis to Generalized Guidance Forms
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2502.07849