AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Semantic Context Matters: Improving Conditioning for Autoregressive Models

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

CVPR 2026 · Image generation › Autoregressive and latent generation · approach: Next-scale and token autoregressive images · 8 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
replace plain resize with learnable semantic compression plus L_pres (4x) (ablation)SCAR · MultiGen-20M DepthFID: 15.78 -> 14.70Improved
add semantic preservation loss L_pres to learnable compression (ablation)SCAR · MultiGen-20M HEDFID: 9.89 -> 9.43Improved
learnable compression versus PixelUnshuffle (ablation)SCAR · MultiGen-20M HEDFID: 9.82 vs 9.89No clear effect
4x compression instead of no compression (ablation)SCAR · MultiGen-20M HEDFID: 9.29 -> 9.43No clear effect
16x compression instead of 4x (ablation)SCAR · MultiGen-20M DepthFID: 14.70 -> 16.10Hurt
use DINOv2-B instead of CLIP-B encoder (ablation)SCAR · MultiGen-20M HEDFID: 13.78 -> 9.43Improved
scale image encoder from DINOv2-S to DINOv2-B (ablation)SCAR · MultiGen-20M DepthFID: 15.26 -> 14.70Improved
use DINOv2-B instead of ViT-B encoder (ablation)SCAR · MultiGen-20M HEDSSIM: 77.45 -> 81.76Improved
add Semantic Alignment Guidance L_align (ablation)SCAR · instruction editing (qualitative)instruction fidelityImproved
raise alignment weight delta to 1.0 (ablation)SCAR · instruction editing (qualitative)visual qualityHurt
use prefilling (SCAR) rather than decoding-stage injection (ControlAR*)LlamaGen-XL · PIE-BenchStructure distance: 116.99 -> 30.98Improved
SCAR versus EditAR on same backboneLlamaGen-XL · PIE-BenchMSE: 130.27 -> 83.47Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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