Search Results for author:
Found 6 papers, 5 papers with code
Self-Corrected Flow Distillation for Consistent One-Step and Few-Step Text-to-Image Generation
Flow matching has emerged as a promising framework for training generative models, demonstrating impressive empirical performance while offering relative ease of training compared to diffusion-based models.
Simple Guidance Mechanisms for Discrete Diffusion Models
Diffusion models for continuous data gained widespread adoption owing to their high quality generation and control mechanisms.
DiMSUM: Diffusion Mamba -- A Scalable and Unified Spatial-Frequency Method for Image Generation
We introduce a novel state-space architecture for diffusion models, effectively harnessing spatial and frequency information to enhance the inductive bias towards local features in input images for image generation tasks.
Flow Matching in Latent Space
In this work, we propose to apply flow matching in the latent spaces of pretrained autoencoders, which offers improved computational efficiency and scalability for high-resolution image synthesis.
Ranked #4 on
Image Generation
on CelebA-HQ 256x256
Anti-DreamBooth: Protecting users from personalized text-to-image synthesis
Despite the complicated formulation of DreamBooth and Diffusion-based text-to-image models, our methods effectively defend users from the malicious use of those models.
Wavelet Diffusion Models are fast and scalable Image Generators
Diffusion models are rising as a powerful solution for high-fidelity image generation, which exceeds GANs in quality in many circumstances.
Ranked #1 on
Image Generation
on CelebA-HQ 512x512
