Improving Consistency Models with Generator-Induced Coupling - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail (Preprint/Prepublication) Année : 2024

Improving Consistency Models with Generator-Induced Coupling

Thibaut Issenhuth
  • Fonction : Auteur
  • PersonId : 1392093
Ludovic dos Santos
Jean-Yves Franceschi
Alain Rakotomamonjy
  • Fonction : Auteur
  • PersonId : 1102415

Résumé

Consistency models are promising generative models as they distill the multi-step sampling of score-based diffusion in a single forward pass of a neural network. Without access to sampling trajectories of a pre-trained diffusion model, consistency training relies on proxy trajectories built on an independent coupling between the noise and data distributions. Refining this coupling is a key area of improvement to make it more adapted to the task and reduce the resulting randomness in the training process. In this work, we introduce a novel coupling associating the input noisy data with their generated output from the consistency model itself, as a proxy to the inaccessible diffusion flow output. Our affordable approach exploits the inherent capacity of consistency models to compute the transport map in a single step. We provide intuition and empirical evidence of the relevance of our generator-induced coupling (GC), which brings consistency training closer to score distillation. Consequently, our method not only accelerates consistency training convergence by significant amounts but also enhances the resulting performance. The code is available at: https://github.com/thibautissenhuth/consistency_GC .
Fichier principal
Vignette du fichier
Generator_induced_consistency_.pdf (1.36 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04611719 , version 1 (13-06-2024)

Identifiants

  • HAL Id : hal-04611719 , version 1

Citer

Thibaut Issenhuth, Ludovic dos Santos, Jean-Yves Franceschi, Alain Rakotomamonjy. Improving Consistency Models with Generator-Induced Coupling. 2024. ⟨hal-04611719⟩
21 Consultations
4 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More