AWhat changed
Organized discrete latent image representations across multiple scales with stronger conditional autoregressive priors to separate global structure and fine detail in image synthesis.
Scientific scope
Organized discrete latent image representations across multiple scales with stronger conditional autoregressive priors to separate global structure and fine detail in image synthesis. Original VQ-VAE is inherited; multi-scale organization/prior integration is the scoped delta, not a new entry merely for more parameters. Likelihood-based generation does not guarantee absence of all diversity failures.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Original VQ-VAE is inherited; multi-scale organization/prior integration is the scoped delta, not a new entry merely for more parameters. Likelihood-based generation does not guarantee absence of all diversity failures.
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.
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