AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

"Principal Components" Enable A New Language of Images

Xin Wen, Bingchen Zhao, Ismail Elezi, Jiankang Deng, Xiaojuan Qi

ICCV 2025 · Image generation › Autoregressive and latent generation · approach: Hierarchical VQ-VAE generation · 13 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 Semanticist DiT-L tokenizer instead of ImageFolderSemanticist · ImageNet 256x256 reconstructionrFID: 0.80 -> 0.72 (DiT-XL: 0.72)Improved
scale diffusion decoder from DiT-L to DiT-XL (ablation)Semanticist · ImageNet 256x256 reconstructionrFID: 0.78 -> 0.72Improved
scale diffusion decoder from DiT-L to DiT-XL (ablation)Semanticist · ImageNet 256x256 reconstructionPSNR: 21.61 -> 21.43Hurt
scale diffusion decoder from DiT-L to DiT-XL (ablation)Semanticist · ImageNet 256x256 reconstructionSSIM: 0.626 -> 0.613Hurt
scale diffusion decoder from DiT-L to DiT-XL (ablation)Semanticist · ImageNet 256x256 class-conditional generationgFID: 2.57 -> 2.57No clear effect
scale diffusion decoder from DiT-L to DiT-XL (ablation)Semanticist · ImageNet 256x256 class-conditional generationIS: 260.9 -> 254.0Hurt
train autoregressive model on first 32 of 256 tokens onlySemanticist · ImageNet 256x256 class-conditional generationgFID: 2.57 with 32 tokens/32 stepsImproved
add REPA regularizer to the DiT decoder (ablation)Semanticist · ImageNet reconstruction (d64x64 tokens)rFIDImproved
add REPA regularizer to the DiT decoder (ablation)Semanticist · ImageNet reconstruction (d64x64 tokens)rFID with all tokensNo clear effect
use lower-dimensional tokens (d16x256 vs d256x16) (ablation)Semanticist · ImageNet reconstructionrFIDImproved
increase CFG scale at inference (ablation)Semanticist · ImageNet reconstruction (smaller model, fewer tokens)reconstruction qualityImproved
use more tokens for linear probing (ablation)Semanticist · ImageNet linear probingtop-1 accuracy: peak 63.5% at few tokens, declining as more tokens addedHurt

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