Starred repositories
🪨 why use many token when few token do trick — Claude Code skill that cuts 65% of tokens by talking like caveman
LucidFlux: Caption-Free Photo-Realistic Image Restoration via a Large-Scale Diffusion Transformer, ICLR 2026
(ICASSP'25) Official Repo for PRIMUS: Pretraining IMU Encoders with Multimodal Self-Supervision
Self-supervised learning for wearables using the UK-Biobank (>700,000 person-days)
[NeurIPS 2025] UniRelight: Learning Joint Decomposition and Synthesis for Video Relighting
A simple video streaming baseline that outperforms SOTAs.
A collection of datasets, papers, and resources for Generalizable Human Activity Recognition and IMU sensing.
[NeurIPS 2025 Spotlight] DiCo: Revitalizing ConvNets for Scalable and Efficient Diffusion Modeling
Emacs extension to interact with the SLURM jobs scheduler
Sharp Monocular View Synthesis in Less Than a Second
[CVPR 2024] Training Generative Image Super-Resolution Models by Wavelet-Domain Losses Enables Better Control of Artifacts
minimal-emacs.d - Emacs init.el and early-init.el for Better Defaults and Optimized Startup, intended to serve as a solid foundation for your vanilla Emacs configuration | Emacs Starter Kit
Python tool for converting files and office documents to Markdown.
The official repository of our ICLR 2026 paper "Vivid-VR: Distilling Concepts from Text-to-Video Diffusion Transformer for Photorealistic Video Restoration".
🔥🔥🔥[AAAI 2026] Official code release of our paper "Fine-grained Image Quality Assessment for Perceptual Image Restoration". 首个面向图像恢复的细粒度IQA数据集FGRestore+方法FGResQ
Tiny AutoEncoder for Stable Diffusion (and other image models)
Nitro-T is a family of text-to-image diffusion models focused on highly efficient training.
attention map tools for huggingface/diffusers
Official Implementation of paper "A Tale of Two Features: Stable Diffusion Complements DINO for Zero-Shot Semantic Correspondence"
Official inference repo for FLUX.1 models
[ICCV 2025] Official implementation of the paper: REPA-E: Unlocking VAE for End-to-End Tuning of Latent Diffusion Transformers
[ICCV2025 highlight]Rectifying Magnitude Neglect in Linear Attention
[ICLR2026 - Oral] WAFT: Warping-Alone Field Transforms for Optical Flow
New flexible and efficient image generation framework that sets new SOTA on FFHQ-256 with FID 2.05, 2022
[NeurIPS'25] One-Step Diffusion for Detail-Rich and Temporally Consistent Video Super-Resolution