Enregistré dans:
Détails bibliographiques
Auteurs principaux: Choi, Changhyun, Kim, Sungha, Kim, H. Jin
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2506.12633
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866918059304812544
author Choi, Changhyun
Kim, Sungha
Kim, H. Jin
author_facet Choi, Changhyun
Kim, Sungha
Kim, H. Jin
contents Recently, it has been shown that investing computing resources in searching for good initial noise for a text-to-image diffusion model helps improve performance. However, previous studies required external models to evaluate the resulting images, which is impossible on GPUs with small VRAM. For these reasons, we apply Best-of-N inference-time scaling to algorithms that optimize the initial noise of a diffusion model without external models across multiple datasets and backbones. We demonstrate that inference-time scaling for text-to-image diffusion models in this setting quickly reaches a performance plateau, and a relatively small number of optimization steps suffices to achieve the maximum achievable performance with each algorithm.
format Preprint
id arxiv_https___arxiv_org_abs_2506_12633
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Performance Plateaus in Inference-Time Scaling for Text-to-Image Diffusion Without External Models
Choi, Changhyun
Kim, Sungha
Kim, H. Jin
Computer Vision and Pattern Recognition
Machine Learning
Recently, it has been shown that investing computing resources in searching for good initial noise for a text-to-image diffusion model helps improve performance. However, previous studies required external models to evaluate the resulting images, which is impossible on GPUs with small VRAM. For these reasons, we apply Best-of-N inference-time scaling to algorithms that optimize the initial noise of a diffusion model without external models across multiple datasets and backbones. We demonstrate that inference-time scaling for text-to-image diffusion models in this setting quickly reaches a performance plateau, and a relatively small number of optimization steps suffices to achieve the maximum achievable performance with each algorithm.
title Performance Plateaus in Inference-Time Scaling for Text-to-Image Diffusion Without External Models
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2506.12633