AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

CSD-VAR: Content-Style Decomposition in Visual Autoregressive Models

Quang-Binh Nguyen, Minh Luu, Quang Nguyen, Anh Tran, Khoi Nguyen

ICCV 2025 · Image generation › Autoregressive and latent generation · approach: Next-scale and token autoregressive images · 5 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
add scale-aware alternating optimization (ablation)CSD-VAR · Validation set (35 concepts)CSD-C: 0.501 -> 0.603Improved
add scale-aware alternating optimization (ablation)CSD-VAR · Validation set (35 concepts)CLIP-T: 0.270 -> 0.330Improved
add SVD-based style rectification (ablation)CSD-VAR · Validation set (35 concepts)DINO: 0.422 -> 0.517Improved
add SVD-based style rectification (ablation)CSD-VAR · Validation set (35 concepts)CSD-S: 0.503 -> 0.564Improved
add augmented K-V memory (ablation)CSD-VAR · Validation set (35 concepts)CLIP-I: 0.702 -> 0.751Improved
add augmented K-V memory (ablation)CSD-VAR · Validation set (35 concepts)CSD-S: 0.559 -> 0.564Improved
apply K-V memory to all 30 blocks instead of the first block (ablation)CSD-VAR · Validation set (35 concepts)CLIP-T: 0.330 -> 0.292Mixed
apply K-V memory to half of the blocks instead of the first block (ablation)CSD-VAR · Validation set (35 concepts)DINO: 0.517 -> 0.550Mixed
use top 10 SVD ranks instead of top 3 (ablation)CSD-VAR · Validation set (35 concepts)CLIP-T: 0.289 -> 0.330Improved
use top 20 SVD ranks instead of top 10 (ablation)CSD-VAR · Validation set (35 concepts)CLIP-T: 0.330 -> 0.302Hurt
use 4 tokens instead of 1 token (ablation)CSD-VAR · Validation set (35 concepts)CSD-C: 0.581 -> 0.603Improved
use 16 tokens instead of 4 tokens (ablation)CSD-VAR · Validation set (35 concepts)CSD-C: 0.603 -> 0.572Hurt

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