Blattmann et al.: temporal alignment of image diffusion and upsamplers

What changed

Inserted and trained temporal components in pretrained image latent-diffusion models and upsamplers, reusing spatial weights for coherent higher-resolution video and personalized-model transfer.

Scientific scope

Inserted and trained temporal components in pretrained image latent-diffusion models and upsamplers, reusing spatial weights for coherent higher-resolution video and personalized-model transfer. Image latent diffusion and upsampling are inherited. Later StableVideoDiffusion curation/staged-training study is distinct. Transfer is tested within compatible model families, not all image generators or physically accurate driving simulation.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Image latent diffusion and upsampling are inherited. Later StableVideoDiffusion curation/staged-training study is distinct. Transfer is tested within compatible model families, not all image generators or physically accurate driving simulation.

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.

• Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Models

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