AWhat changed
Combined a perceptual/adversarially trained discrete image autoencoder with transformer modeling of latent image tokens for conditional high-resolution synthesis.
Scientific scope
Combined a perceptual/adversarially trained discrete image autoencoder with transformer modeling of latent image tokens for conditional high-resolution synthesis. Vector quantization, GANs and transformers retain prior credit; acquired delta is their perceptual compression/latent modeling integration. Original preprint is2020 and CVPR publication2021. Not the later latent diffusion architecture.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Vector quantization, GANs and transformers retain prior credit; acquired delta is their perceptual compression/latent modeling integration. Original preprint is2020 and CVPR publication2021. Not the later latent diffusion architecture.
Contributors
- Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.
- Taming Transformers for High-Resolution Image Synthesis
- Esser et al.: perceptual adversarial image tokens with transformer modeling — author/institutional project
Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction