AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Boosting Resolution Generalization of Diffusion Transformers with Randomized Positional Encodings

Liang Hou, Cong Liu, Mingwu Zheng, Xin Tao, Pengfei Wan, Di Zhang, Kun Gai

AAAI 2026 · Image generation › Autoregressive and latent generation · approach: High-resolution diffusion synthesis · 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
add random resize-and-crop augmentation with micro-conditioning (ablation)RPE-2D DiT-XL/2 · ImageNet 512x512 extrapolationFID: 20.78 -> 19.12Improved
add random resize-and-crop augmentation with micro-conditioning (ablation)RPE-2D DiT-XL/2 · ImageNet 512x512 extrapolationIS: 293.96 -> 325.76Improved
add attention scaling at inference (ablation)RPE-2D DiT-XL/2 · ImageNet 512x512 extrapolationsFID: 23.91 -> 19.17Improved
add timestep shifting at inference (ablation)RPE-2D DiT-XL/2 · ImageNet 512x512 extrapolationFID: 18.33 -> 17.95Improved
replace YaRN with RPE-2D for extrapolation 256->512 (ablation)DiT-XL/2 (RoPE) · ImageNet 512x512 (trained at 256)sFID: 34.31 -> 18.23Improved
replace NTK with RPE-2D for extrapolation 512->1024 (ablation)DiT-XL/2 (RoPE) · ImageNet 1024x1024 (trained at 512)FID: 32.90 -> 25.40Improved
apply RPE-2D to RoPE (ablation)DiT-XL/2 · ImageNet 256x256 in-distributionFID: 17.30 -> 16.92Improved
apply RPE-2D to sinusoidal PE (ablation)DiT-XL/2 · ImageNet 256x256 in-distributionFID: 17.82 -> 17.33Improved
compare RPE-2D to PI on precision at 768x768DiT-XL/2 · ImageNet 512x512 (trained at 512)Precision: PI precision 0.8441 vs RPE-2D 0.8271 at 512Mixed
use RPE-2D when fine-tuning 256->512DiT-XL/2 RoPE · 512 fine-tuningtraining lossImproved

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