Official Pytorch implementation of "E2VPT: An Effective and Efficient Approach for Visual Prompt Tuning". (ICCV2023)

A GPU-accelerated library containing highly optimized building blocks and an execution engine for data processing to accelerate deep learning training and inference applications.

[NeurIPS'22] This is an official implementation for "Scaling & Shifting Your Features: A New Baseline for Efficient Model Tuning".

Train latent diffusion for real-world super-resolution.

Fine-tune SAM (Segment Anything Model) for computer vision tasks such as semantic segmentation, matting, detection ... in specific scenarios

[ICML2024 (Oral)] Official PyTorch implementation of DoRA: Weight-Decomposed Low-Rank Adaptation

[ICLR 2025] Diffusion Feedback Helps CLIP See Better

Official Pytorch Implementation of Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think (ICLR 2025)