Code for MIRO: MultI-Reward cOnditioned pretraining improves T2I quality and efficiency

DiffusionNFT: Online Diffusion Reinforcement with Forward Process

Lyra: Generative 3D Scene Reconstruction via Video Diffusion Model Self-Distillation

[ICCV2025]Generate one 2K image on single 3090 GPU!

[ICCV 2025] FoundIR: Unleashing Million-scale Training Data to Advance Foundation Models for Image Restoration

#ICCV, #MoE, #Tracking

#ICCV 2025, # FlexTrack,# Mixture of Experts

An official implementation of DanceGRPO: Unleashing GRPO on Visual Generation

[NeurIPS 2025] An official implementation of Flow-GRPO: Training Flow Matching Models via Online RL

[CVPR 2025] Official PyTorch implementation of Not All Parameters Matter: Masking Diffusion Models for Enhancing Generation Ability

[ICML2025] VARSR: Visual Autogressive Modeling for Image Super Resolution

Paper List of Inference/Test Time Scaling/Computing

The official implementation of "2025ICLR Dynamic Diffusion Transformer" and "2025ArXivDyDiT++: Dynamic Diffusion Transformers for Efficient Visual Generation".

[CVPR 2025 Highlight] TinyFusion: Diffusion Transformers Learned Shallow

Scene-Centric Unsupervised Panoptic Segmentation (CVPR 2025 Highlight)

Official inference repo for FLUX.1 models

codes for RFSR: Improving ISR Diffusion Models via Reward Feedback Learning

[ICCV'25] Compression-Aware One-Step Diffusion Model for JPEG Artifact Removal

[ECCV2024] The Official Code for "RealViformer: Investigating Attention for Real-World Video Super-Resolution"

Official repository of "SAMURAI: Adapting Segment Anything Model for Zero-Shot Visual Tracking with Motion-Aware Memory"

[ECCV 2024] OneRestore: A Universal Restoration Framework for Composite Degradation

A generative world for general-purpose robotics & embodied AI learning.

PhysGen: Rigid-Body Physics-Grounded Image-to-Video Generation (ECCV 2024)

[NeurIPS 2024] Official code for "Splatter a Video: Video Gaussian Representation for Versatile Processing"

[ICLR'25 Oral] Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think

[NeurIPS 2024] NeuMA: Neural Material Adaptor for Visual Grounding of Intrinsic Dynamics

[NeurIPS 24] PromptFix: You Prompt and We Fix the Photo

[NeurIPS 2024 Best Paper Award][GPT beats diffusion🔥] [scaling laws in visual generation📈] Official impl. of "Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction". A…

Repository for "Enhanced Super-Resolution Training via Mimicked Alignment for Real-World Scenes", ACCV 2024

SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformer