AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Exploring Position Encoding Mechanism in Diffusion U-Net for Training-free High-resolution Image Generation

Feng Zhou, Pu Cao, Yiyang Ma, Lu Yang, Yonghao Dang, Jianqin Yin

AAAI 2026 · Image generation › Autoregressive and latent generation · approach: High-resolution diffusion synthesis · 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
replace direct inference with PBCSD-XL · 2048x2048 generationKID: 0.0092 -> 0.0067Improved
replace direct inference with PBCSD-XL · 2048x2048 generationContent Richness: 30.30 -> 29.42Improved
replace direct inference with PBCSD-XL · 2048x2048 generationImageReward: 0.645 -> 1.036Improved
replace direct inference with PBCSD-XL · 2048x2048 generationInference time (s): 68s -> 71s (+4%)Hurt
use PBC instead of ScaleCrafterSD-XL · 2048x2048 generationIS: 9.53 -> 9.95Improved
use PBC instead of FouriScaleSD-XL · 2048x2048 generationInference time (s): 130s (+91%) -> 71s (+4%)Improved
use PBC instead of ScaleCrafterSD-XL · 2048x2048 generationImageReward: 1.078 -> 1.036Hurt
add PBC virtual boundaries to zero-padding U-Net (ablation)SD-2.1 · position loss, 1024^2Linear-probe position loss: 0.023 -> 0.017Improved
add PBC virtual boundaries to zero-padding U-Net (ablation)SD-2.1 · position loss, central 512^2 in 1024^2Linear-probe position loss: 0.039 -> 0.010Improved
use PBC instead of dilated convolution (ablation)SD-2.1 · position loss, central 512^2 in 1024^2Linear-probe position loss: 0.025 -> 0.010Improved
add one virtual boundary (N=0 to N=1) (ablation)PBC · position lossLinear-probe position loss: 0.0230 -> 0.0173Improved
increase virtual boundaries from N=1 to N=7 (ablation)PBC · position lossLinear-probe position loss: 0.0173 -> 0.0175No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 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