taohu/zigma
Text-to-Image
•
Updated
•
2
Tao Hu
taohu
AI & ML interests
None yet
Recent Activity
updated
a model
about 2 months ago
taohu/zigma
New activity
about 2 months ago
taohu/zigma:Add link to paper
Organizations
None yet
taohu's activity
updated
a
model
about 2 months ago
Add link to paper
#1 opened about 2 months ago
by
nielsr
nielsr
updated
a
dataset
2 months ago
liked
a
dataset
2 months ago
AttributeError: module 'PIL.Image' has no attribute 'ExifTags'.
1
#2 opened 3 months ago
by
geekyrakshit
geekyrakshit
updated
a
model
2 months ago
updated
a
dataset
2 months ago
upvoted
a
paper
4 months ago
upvoted
an
article
6 months ago
Article
Merge Large Language Models with mergekit
By
•
•
80
liked
a
Space
6 months ago
liked
a
model
6 months ago
updated
a
model
6 months ago
liked
a
dataset
7 months ago
liked
a
model
7 months ago
Reacted to
akhaliq's
post with 🔥
8 months ago
Post
2219
Fast High-Resolution Image Synthesis with Latent Adversarial Diffusion Distillation
Fast High-Resolution Image Synthesis with Latent Adversarial Diffusion Distillation (2403.12015)
Diffusion models are the main driver of progress in image and video synthesis, but suffer from slow inference speed. Distillation methods, like the recently introduced adversarial diffusion distillation (ADD) aim to shift the model from many-shot to single-step inference, albeit at the cost of expensive and difficult optimization due to its reliance on a fixed pretrained DINOv2 discriminator. We introduce Latent Adversarial Diffusion Distillation (LADD), a novel distillation approach overcoming the limitations of ADD. In contrast to pixel-based ADD, LADD utilizes generative features from pretrained latent diffusion models. This approach simplifies training and enhances performance, enabling high-resolution multi-aspect ratio image synthesis. We apply LADD to Stable Diffusion 3 (8B) to obtain SD3-Turbo, a fast model that matches the performance of state-of-the-art text-to-image generators using only four unguided sampling steps. Moreover, we systematically investigate its scaling behavior and demonstrate LADD's effectiveness in various applications such as image editing and inpainting.
Fast High-Resolution Image Synthesis with Latent Adversarial Diffusion Distillation (2403.12015)
Diffusion models are the main driver of progress in image and video synthesis, but suffer from slow inference speed. Distillation methods, like the recently introduced adversarial diffusion distillation (ADD) aim to shift the model from many-shot to single-step inference, albeit at the cost of expensive and difficult optimization due to its reliance on a fixed pretrained DINOv2 discriminator. We introduce Latent Adversarial Diffusion Distillation (LADD), a novel distillation approach overcoming the limitations of ADD. In contrast to pixel-based ADD, LADD utilizes generative features from pretrained latent diffusion models. This approach simplifies training and enhances performance, enabling high-resolution multi-aspect ratio image synthesis. We apply LADD to Stable Diffusion 3 (8B) to obtain SD3-Turbo, a fast model that matches the performance of state-of-the-art text-to-image generators using only four unguided sampling steps. Moreover, we systematically investigate its scaling behavior and demonstrate LADD's effectiveness in various applications such as image editing and inpainting.
upvoted
a
collection
8 months ago
upvoted
2
papers
8 months ago
authored
2
papers
8 months ago