-
The Illustrated Stable Diffusion – Jay Alammar – Visualizing machine learning one concept at a time.
-
Lesson 9: Deep Learning Foundations to Stable Diffusion, 2022 - YouTube
-
GitHub - huggingface/diffusion-models-class: Materials for the Hugging Face Diffusion Models Course
-
Sana - Efficient High-Resolution Image Synthesis with Linear Diffusion Transformer
-
[2410.07815] Simple ReFlow: Improved Techniques for Fast Flow Models
-
[2410.10733v1] Deep Compression Autoencoder for Efficient High-Resolution Diffusion Models
-
23672 - 2nd Workshop on Generative Models for Computer Vision - YouTube
-
GitHub - openai/consistency_models: Official repo for consistency models.
-
Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think
-
textual inversion
- optimize a new token embedding for new concept (an image is worth one word)
-
ControlNet
- Copies the model and grafts it onto existing frozen one and tunes it to condition on things like depth or edge maps
-
DreamBooth
- find unused tokens to map the concept onto, update (fine tune) the model
- auxiliary loss to penalize drift for other concepts
-
LoRA
- zipLoRA
- C-LoRA
Unlearning
- optimize to maximize loss for given concept