AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

SCALAR: Scale-wise Controllable Visual Autoregressive Learning

Ryan Xu, Dongyang Jin, Yancheng Bai, Rui Lan, Xu Duan, Lei Sun, Xiangxiang Chu

AAAI 2026 · Image generation › Autoregressive and latent generation · approach: Next-scale and token autoregressive images · 17 citations (OpenAlex)

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
use DINOv2-S encoder instead of ViT-S (ablation)SCALAR VAR-d12 · ImageNet-256 CannyFID: 4.34 -> 3.12Improved
use DINOv2-S encoder instead of SAM-B (ablation)SCALAR VAR-d12 · ImageNet-256 CannyFID: 6.13 -> 3.12Improved
use larger DINOv2-B encoder instead of DINOv2-S (ablation)SCALAR VAR-d12 · ImageNet-256 DepthFID: 3.83 -> 3.68Improved
scale VAR backbone from d12 to d24 (ablation)SCALAR · ImageNet-256 CannyFID: 2.95 -> 2.14Improved
add Unified Control Alignment loss (ablation)SCALAR-Uni-d12 · ImageNet-256 CannyFID: 3.73 -> 3.31Improved
add Unified Control Alignment loss (ablation)SCALAR-Uni-d20 · ImageNet-256 DepthRMSE: 36.88 -> 37.12Hurt
add Unified Control Alignment loss (ablation)SCALAR-Uni-d20 · ImageNet-256 CannyF1-Score: 31.5 -> 31.7No clear effect
unify five conditions in one model (SCALAR-Uni vs SCALAR)SCALAR-Uni-d12 · ImageNet-256 DepthFID: 3.83 -> 4.21Hurt
freeze all VAR backbone parameters (ablation)SCALAR · ImageNet-256control performanceHurt
use SCALAR-d12 instead of ControlAR LlamaGen-LSCALAR VAR-d12 · ImageNet-256 CannyFID: 7.69 -> 3.12Improved
SCALAR-d24 vs ControlAR LlamaGen-L control consistencySCALAR VAR-d24 · ImageNet-256 CannyF1-Score: 34.9 vs 33.1Hurt
SCALAR-d24 vs ControlAR LlamaGen-L control consistencySCALAR VAR-d24 · ImageNet-256 DepthRMSE: 31.1 vs 33.13Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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