SVG-T2I: Scaling Up Text-to-Image Latent Diffusion Model Without Variational Autoencoder
Paper
•
2512.11749
•
Published
•
36
Haolin Wang
howlin
AI & ML interests
multimodal generation, unified vision model
Recent Activity
liked
a dataset
2 days ago
timm/imagenet-w21-wds
liked
a model
17 days ago
howlin/SVG
authored
a paper
17 days ago
SVG-T2I: Scaling Up Text-to-Image Latent Diffusion Model Without Variational Autoencoder
Organizations
None yet
authored
a
paper
17 days ago
authored
3
papers
2 months ago
DC-Solver: Improving Predictor-Corrector Diffusion Sampler via Dynamic Compensation
Paper
•
2409.03755
•
Published
•
4
InstaRevive: One-Step Image Enhancement via Dynamic Score Matching
Paper
•
2504.15513
•
Published
Latent Diffusion Model without Variational Autoencoder
Paper
•
2510.15301
•
Published
•
49