AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Diffusion Image Generation with Explicitly Modeling of Data Manifold Geometry

Duoduo Xue, Zhiyu Zhu, Junhui Hou

ECCV 2026 · Image generation › Autoregressive and latent generation · approach: High-resolution diffusion synthesis

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 DiT-B/2 with MIND-B (discrete-token hypersphere diffusion)DiT-B/2 · ImageNet 256x256 (80 epochs, no guidance)FID: 43.47 -> 22.73Improved
replace DiT-S/2 with MIND-SDiT-S/2 · ImageNet 256x256 (80 epochs, no guidance)FID: 68.40 -> 40.72Improved
replace SiT-B/2 with MIND-BSiT-B/2 · ImageNet 256x256 (80 epochs, no guidance)FID: 33.02 -> 22.73Improved
replace SiT-B/2 with MIND-BSiT-B/2 · ImageNet 256x256 (80 epochs, no guidance)Recall: 0.63 -> 0.58Hurt
replace DiT-S/2 with MIND-S under guidance cfg 1.5DiT-S/2 · ImageNet 256x256 (80 epochs, cfg 1.5)FID: 44.42 -> 22.33Improved
replace SiT-S/2 with MIND-S under guidance cfg 1.5SiT-S/2 · ImageNet 256x256 (80 epochs, cfg 1.5)FID: 35.62 -> 22.33Improved
replace DiT-B/2 with MIND-B under guidance cfg 2.0DiT-B/2 · ImageNet 256x256 (80 epochs, cfg 2.0)FID: 10.95 -> 7.97Improved
replace SiT-B/2 with MIND-B under guidance cfg 2.0SiT-B/2 · ImageNet 256x256 (80 epochs, cfg 2.0)FID: 9.07 -> 7.97Improved
replace DiT-XL/2 with MIND-XL-G in system-level comparisonDiT-XL/2 · ImageNet 256x256 (system-level, with guidance)FID: 2.27 -> 1.95Improved
use stronger GigaTok tokenizer and longer training for MIND-BMIND-B · ImageNet 256x256 (system-level, with guidance)FID: 2.18 -> 2.06Improved
use MIND-B (130M) instead of LlamaGen-3B (3.1B)LlamaGen-3B · ImageNet 256x256 (system-level, with guidance)FID: 2.18 -> 2.06Improved
use MIND-S instead of eMIGM-S at one-third of the parameterseMIGM-S · ImageNet 256x256 (80 epochs, no guidance)FID: 44.47 -> 40.72Improved

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