AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Anomaly-Preference Image Generation

Fuyun Wang, Yuanzhi Wang, Xu Guo, Sujia Huang, Tong Zhang, Dan Wang, Xin Liu, Hui Yan, Zhen Cui

ICML 2026 · Image generation › Fast diffusion sampling · approach: Distributional control of diffusion · 0 citations (OpenAlex)

Read the paper

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 CPO implicit preference alignment (ablation)APO · MVTec capsule/grid/wood localizationAUC-P / AP-P / F1-PImproved
add Time-Aware Capacity Allocation (TACA) (ablation)APO · MVTec capsule/grid/wood localizationAUC-P / AP-P / F1-PImproved
set TACA minimum rank kmin to 4 (versus 1, 2, 8, 16) (ablation)APO · MVTec generation (IS, IC-LPIPS)IS / IC-LPIPSMixed
use implicit preference alignment with TACA and hierarchical sampling instead of prior generatorsAPO · MVTec generation qualityaverage IS: 2.02 -> 2.14Improved
train U-Net on APO synthetic anomalies instead of baseline generatorsU-Net detector · MVTec AD pixel-levelAUC-P: 99.1 -> 99.6Improved
train U-Net on APO synthetic anomalies instead of AnomalyDiffusion dataU-Net detector · MVTec AD image-levelAP-I: 99.7 -> 99.5No clear effect
train ResNet-34 classifier on APO synthetic anomaliesResNet-34 · MVTec anomaly classificationaccuracy (%): 70.25 -> 72.06Improved
use U-Net trained on APO data instead of supervised detectorsanomaly localization · MVTec ADAUROC/AP average: 99.0/78.6 -> 99.6/87.7 (vs PRN)Improved

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