Editing Implicit Assumptions in Text-to-Image Diffusion Models
Paper
β’
2303.08084
β’
Published
β’
2
This is a forked version of the diffusers π€ implementation of Text-to-image model editing (TIME, Editing Implicit Assumptions in Text-to-Image Diffusion Models).
The code in this repository is managed at py-img-gen/diffusers-text-to-model-editing.
Here are the minor changes:
CrossAttention class to the Attention classrestart_params with large side effects from the edit_model function