AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

AccDiffusion v2: Toward More Accurate Higher-Resolution Diffusion Extrapolation

Zhihang Lin, Mingbao Lin, Wengyi Zhan, Rongrong Ji

TPAMI 2025 · Image generation › Autoregressive and latent generation · approach: High-resolution diffusion synthesis · 3 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
upgrade AccDiffusion to AccDiffusion v2 (patch-aware prompts, dilated sampling with interaction, ControlNet)AccDiffusion · 4096x4096 (16x)FID_r: 63.89 -> 60.88Improved
upgrade AccDiffusion to AccDiffusion v2AccDiffusion · 2048x2048 (4x)IS_r: 16.48 -> 18.62Improved
upgrade AccDiffusion to AccDiffusion v2AccDiffusion · 3072x3072 (9x)FID_c: 46.46 -> 44.90Improved
upgrade AccDiffusion to AccDiffusion v2AccDiffusion · 4096x4096 (16x)CLIP score: 33.79 -> 32.83Hurt
upgrade AccDiffusion to AccDiffusion v2AccDiffusion · 4096x4096 (16x)inference time (min): 26 min -> 35 minHurt
remove patch-content-aware prompts (ablation)AccDiffusion v2 · 4096x4096 (16x)qualitativeImproved
remove dilated sampling with window interaction (ablation)AccDiffusion v2 · 4096x4096 (16x)qualitativeImproved
remove ControlNet-assisted generation (ablation)AccDiffusion v2 · 4096x4096 (16x)qualitativeImproved
increase ControlNet steps (ablation)AccDiffusion v2 · 4096x4096 (16x)image quality vs timeMixed
use very small patch prompt threshold c (ablation)AccDiffusion v2 · 4096x4096 (16x)qualitativeHurt
use very large patch prompt threshold c (ablation)AccDiffusion v2 · 4096x4096 (16x)qualitativeHurt
run on H100 instead of RTX 3090AccDiffusion v2 · 4096x4096 (16x)inference time (min): 35 min -> 11 minImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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