AI Paper Atlas

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

Wasserstein-Aware Transfer: Class-Level Alignment for Robust Diffusion Model Adaptation

Zixian Huang, Chuan-Xian Ren

AAAI 2026 · Image generation › Text-to-image diffusion · approach: Multi-concept composition · 0 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 WAT (class matching + class-level retention sampling) instead of CFG after fine-tuningDiT-XL/2-256 fine-tuning · Seven downstream datasets averageFID: 13.69 -> 10.73Improved
use WAT instead of Domain guidanceDiT-XL/2-256 fine-tuning · Seven downstream datasets averageFID: 11.55 -> 10.73Improved
use WAT instead of Domain guidanceDiT-XL/2-256 fine-tuning · Seven downstream datasets averageFDDINOv2: 292.62 -> 257.71Improved
use WAT instead of Domain guidanceDiT-XL/2-256 fine-tuning · ArtBench-10 (large domain shift)FID: 16.76 -> 13.43Improved
use WAT instead of Domain guidanceDiT-XL/2-256 fine-tuning · CUB, Stanford Cars, Caltech-101FID: 3.52 -> 3.55 (CUB); 4.38 -> 4.43 (Cars); 23.05 -> 23.40 (Caltech)Mixed
restrict fine-tuning timestep sampling to [0, 600] (ablation)DiT fine-tuning with CFG · Seven downstream datasetsFID: Food -2.84%, SUN +5.52%, Caltech -4.82%, CUB +6.70%, Cars +5.38%, DF20 +13.54%, ArtBench +1.15%Mixed
use pretrained model for noisy steps [600,1000] and fine-tuned model for the rest at sampling (ablation)Hybrid sampling · Stanford Cars and Food-101 samplesqualitative generationImproved
use WAT at fewer sampling steps (25) instead of CFG (ablation)WAT sampling · SUN397FID: 24.34 -> 15.66Improved
use WAT at 100 sampling steps instead of DoG (ablation)WAT sampling · CUB-200-2011FID: 3.35 -> 3.66Hurt
vary CLR scale w away from 0.5 (ablation)WAT · Food, CUB, CarFIDHurt

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