Generative Semantic Nursing
2 papers with code • 0 benchmarks • 0 datasets
Generative Semantic Nursing is a task of intervening the generative process on the fly during inference time to improve the faithfulness of the generated images. It works by carefully manipulating of latents during the denoising process of a pre-trained text-to-image diffusion model.
Source: Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Models
Benchmarks
These leaderboards are used to track progress in Generative Semantic Nursing
Most implemented papers
Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Models
AttendAndExcite/Attend-and-Excite
•
•
Recent text-to-image generative models have demonstrated an unparalleled ability to generate diverse and creative imagery guided by a target text prompt.
Divide & Bind Your Attention for Improved Generative Semantic Nursing
boschresearch/Divide-and-Bind
•
•
To address the challenges posed by complex prompts or scenarios involving multiple entities and to achieve improved attribute binding, we propose Divide & Bind.
