AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

Att-Adapter: A Robust and Precise Domain-Specific Multi-Attributes T2I Diffusion Adapter via Conditional Variational Autoencoder

Wonwoong Cho, Yan-Ying Chen, Matthew Klenk, David I. Inouye, Yanxia Zhang

ICCV 2025 · Image generation › Text-to-image diffusion · approach: Compositional text-to-image guidance · 1 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
train Att-Adapter on real unpaired multi-attribute data instead of using GAN-derived paired baselinesAtt-Adapter · FFHQ latent control, single attributecontrol range (CR): S-I-GANs 49.45 -> 67.87Improved
joint multi-attribute training for disentanglementAtt-Adapter · FFHQ latent control, multi attributeDIS: 20.1% -> 32.3%Improved
use continuous attribute embedding instead of discretized LoRA tokensAtt-Adapter · FFHQ absolute controlfacial composition L1: 17.48 -> 15.57Improved
use continuous attribute embedding instead of discretized LoRA tokens on EVOXAtt-Adapter · EVOX size controlsize accuracy: 67% -> 95%Improved
add CVAE to the adapter (ablation)Att-Adapter · FFHQ fixed attribute conditionID similarity: 28.5% -> 6.2%Improved
add CVAE (ablation)Att-Adapter · FFHQ fixed attribute conditioncontrol range (CR): 78.27 -> 78.59No clear effect
train naive adapter longer (20k to 200k) (ablation)Att-Adapter w/o CVAE · FFHQCR and ID similarity: CR 47.04 -> 78.27; ID Sim 6.0% -> 28.5%Mixed
use dropout 0.25 instead of CVAE for regularization (ablation)naive Att-Adapter · FFHQID similarity: CVAE 6.2% vs dropout 23.6%Hurt
increase attribute strength lambda from 0.0 to 0.9 (ablation)Att-Adapter · FFHQfacial composition L1 and ChatGPT alignment: L1 69.43 -> 15.57; ChatGPT 97% -> 78%Mixed
use non-isotropic instead of isotropic Gaussian prior (ablation)Att-Adapter · FFHQfacial composition L1No clear effect
increase latent dimension Z from 128 to 512 (ablation)Att-Adapter · FFHQfacial composition L1: 20.54 -> 17.42Improved
compare ITI-GEN (naive) against Att-Adapter for 20 attributesITI-GEN · FFHQ absolute controlfacial composition L1: 15.57 -> 52.16Hurt

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