Tooth and Claw

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Hierarchical Text-Conditional Image Generation with CLIP Latents (2022, methodology): “[19, 36, 32, 18] guide GANs using gradients from a CLIP model.”
GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models (2021, neutral): “Since CLIP provides a score of how close an image is to a caption, several works have used it to steer generative models like GANs towards a user-deﬁned text caption (Galatolo et al., 2021; Patashnik et al., 2021; Murdock, 2021; Gal et al., 2021).”
Paint by Word (2021, background): “Artist Murdock has observed that CLIP can be used as a source of gradients to guide a generator [29, 28, 30].”
The Creativity of Text-to-Image Generation (2022, background): “Ryan Murdoch created “The Big Sleep” [5, 32], a combination of a GAN called BigGAN and CLIP.”
An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion (2022, background): “Rather than training conditional models, several approaches employ test-time optimization to explore the latent spaces of a pre-trained generator (Crowson et al., 2022; Murdock, 2021; Crowson, 2021).”