AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Hyperspherical Latents Improve Continuous-Token Autoregressive Generation

Guolin Ke, HUI XUE

ICLR 2026 · Image generation › Autoregressive and latent generation · approach: Accelerating autoregressive image generation

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add normalization of AR inputs and outputs (after decoder-input normalization) (ablation)SphereAR-L with Gaussian VAE · ImageNet 256x256 (50-epoch ablation)FID: 2.89 -> 2.68Improved
add normalization only on VAE decoder input (ablation)SphereAR-L with Gaussian VAE · ImageNet 256x256 (50-epoch ablation)FID: 2.97 -> 2.89Improved
use spherical (Power Spherical) posterior instead of Gaussian with normalization (ablation)SphereAR-L · ImageNet 256x256 (50-epoch ablation)FID: 2.68 -> 2.52Improved
use S-VAE instead of MAR's VAE (ablation)SphereAR-L · ImageNet 256x256 (50-epoch ablation)FID: 4.54 -> 2.52Improved
fix the posterior variance (sigma-VAE) instead of learned diagonal Gaussian (ablation)AR with VAE latents · ImageNet 256x256 (50-epoch ablation)FID vs CFGNo clear effect
increase KL weight of diagonal-Gaussian VAE (ablation)AR with Gaussian VAE · ImageNet 256x256 (50-epoch ablation)FID vs CFGMixed
apply post-hoc normalization to diagonal-Gaussian latents (ablation)AR with Gaussian VAE · ImageNet 256x256 (50-epoch ablation)FID vs CFGImproved
use hybrid CNN+ViT VAE backbone instead of pure CNN (ablation)S-VAE · ImageNet 256x256 (50-epoch ablation)training speed and FID: speed 2.25 -> 5.81 it/s; FID 2.63 -> 2.52Improved
use pure ViT VAE backbone instead of hybrid (ablation)S-VAE · ImageNet 256x256 (50-epoch ablation)FID: 2.52 -> 2.81Hurt
reduce diffusion-head sampling steps (100 to 10 to 1) (ablation)SphereAR · inference speed on A100inference speedImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy