MPDiT: Multi-Patch Global-to-Local Transformer Architecture For Efficient Flow Matching and Diffusion Model
Paper • 2603.26357 • Published • 4
Quan Dao
quandao10
AI & ML interests
Generative Model, Computer Vision, Machine Learning
Recent Activity
authored a paper 10 days ago
MPDiT: Multi-Patch Global-to-Local Transformer Architecture For Efficient Flow Matching and Diffusion Model submitted a paper 10 days ago
MPDiT: Multi-Patch Global-to-Local Transformer Architecture For Efficient Flow Matching and Diffusion ModelOrganizations
authored a
paper 10 days ago
submitted a
paper to Daily Papers 10 days ago
authored a
paper 13 days ago
authored 2
papers 7 months ago
authored 3
papers about 1 year ago
DiMSUM: Diffusion Mamba -- A Scalable and Unified Spatial-Frequency Method for Image Generation
Paper • 2411.04168 • Published • 4
Self-Corrected Flow Distillation for Consistent One-Step and Few-Step Text-to-Image Generation
Paper • 2412.16906 • Published • 2
Improved Training Technique for Latent Consistency Models
Paper • 2502.01441 • Published • 8
authored a
paper over 1 year ago
authored 4
papers about 2 years ago
Wavelet Diffusion Models are fast and scalable Image Generators
Paper • 2211.16152 • Published • 1
Flow Matching in Latent Space
Paper • 2307.08698 • Published • 1
Anti-DreamBooth: Protecting users from personalized text-to-image synthesis
Paper • 2303.15433 • Published • 3
Robust Diffusion GAN using Semi-Unbalanced Optimal Transport
Paper • 2311.17101 • Published • 2