xLeaF Lab
Pinned Loading
Repositories
- flash-cuda-mode Public
xie-lab-ml/flash-cuda-mode’s past year of commit activity - awesome-alignment-of-diffusion-models Public
The collection of awesome papers on alignment of diffusion models.
xie-lab-ml/awesome-alignment-of-diffusion-models’s past year of commit activity - Golden-Noise-for-Diffusion-Models Public
[ICCV2025] The code of our work "Golden Noise for Diffusion Models: A Learning Framework".
xie-lab-ml/Golden-Noise-for-Diffusion-Models’s past year of commit activity - multiphysics-bench Public
We propose the first Multiphysics Bench for benchmarking and investigating Scientific Machine Learning for solving multiphysics PDEs.
xie-lab-ml/multiphysics-bench’s past year of commit activity - flow-inference-enhancement Public
We propose a novel training-free inference enhancement method for flow-based diffusion models.
xie-lab-ml/flow-inference-enhancement’s past year of commit activity - Weak-to-Strong-Diffusion-with-Reflection Public
The official code of "Weak-to-Strong Diffusion with Reflection".
xie-lab-ml/Weak-to-Strong-Diffusion-with-Reflection’s past year of commit activity - CoRe2 Public
The official implementation of our paper "CoRe^2: Collect, Reflect and Refine to Generate Better and Faster".
xie-lab-ml/CoRe2’s past year of commit activity - Zigzag-Diffusion-Sampling Public
[ICLR2025] The code of Z-Sampling, proposed in our paper "Zigzag Diffusion Sampling: Diffusion Models Can Self-Improve via Self-Reflection".
xie-lab-ml/Zigzag-Diffusion-Sampling’s past year of commit activity - IV-mixed-Sampler Public
[ICLR2025] IV-Mixed Sampler: Leveraging Image Diffusion Models for Enhanced Video Synthesis
xie-lab-ml/IV-mixed-Sampler’s past year of commit activity - Meissonic-Inference Public
Bag of Design Choices for Inference of High-Resolution Masked Generative Transformer
xie-lab-ml/Meissonic-Inference’s past year of commit activity
People
This organization has no public members. You must be a member to see who’s a part of this organization.
Most used topics
Loading…