XIANG Weilai
XIANG Weilai
## ❓ Questions Hello, I wonder if I can manipulate the textures in the scene? For example, **changing the texture mapping image** of an object to add some perturbation on...
https://github.com/labmlai/annotated_deep_learning_paper_implementations/blob/05632f9f8e0de4657c210a13954a81f9556fd1ed/labml_nn/diffusion/ddpm/unet.py#L188 According to my understanding of Self-Attention, the softmax operation should be done along the `j` axis in einsum? https://github.com/labmlai/annotated_deep_learning_paper_implementations/blob/05632f9f8e0de4657c210a13954a81f9556fd1ed/labml_nn/diffusion/ddpm/unet.py#L190 **So, I think the code should be `attn = attn.softmax(dim=2)`.**...
Paper name: Denoising Diffusion Autoencoders are Unified Self-supervised Learners Paper link: https://arxiv.org/abs/2303.09769 Code link: https://github.com/FutureXiang/ddae Related area: Diffusion, Self-supervised representation learning Selected for *Oral* presentation at ICCV 23. Thank you!
Hi, I try to train the EDM model with a simpler 35.7M #params UNet (proposed by original DDPM paper) and compare the result with DDPM/DDIM. I notice that $S_{churn} =...
Dear authors, I'm really curious about the efficiency of the proposed DiffiT models. It seems that another concurrent work from NVIDIA (by Karras), namely [Analyzing and Improving the Training Dynamics...
Dear authors, Thank you for sharing this amazing and inspiring work! I am really interested in the CIFAR-100 experiment presented in your paper, and I have a couple of questions...