AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

MMDiff: Extending Diffusion Transformers for Multi-Modal Generation

Yagmur Akarken, Orest Kupyn, Christian Rupprecht

ECCV 2026 · Image generation › Text-to-image diffusion · approach: Compositional text-to-image guidance · 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 multi-timestep aggregation (4 timesteps) over single-timestep t=0 extraction (ablation)MMDiff (single-stream + concept attention) · PASCAL VOC 2012mIoU: 50.20 -> 78.90 (+28.7%)Improved
use single-stream blocks instead of double-stream blocks (t=0) (ablation)MMDiff · PASCAL VOC 2012mIoU: 40.52 -> 50.20Improved
add concept attention to single-timestep features (ablation)MMDiff · PASCAL VOC 2012mIoU: 50.20 -> 63.77 (+13.57%)Improved
use concept attention instead of averaged cross-attention maps (ablation)MMDiff (single timestep) · PASCAL VOC 2012mIoU: 58.31 -> 63.77Improved
add cross-attention maps to single-stream t=0 features (ablation)MMDiff · DUTSS-measure: 0.890 -> 0.869Hurt
add DINO-v3 features to multi-timestep FLUX features (ablation)MMDiff · PASCAL VOC 2012mIoU: 78.90 -> 84.95Improved
add spatially varying aggregation weights to full model (ablation)MMDiff + DINO-v3 · NYU Depth v2AbsRel: 0.130 -> 0.116Improved
add aggregation to full model (ablation)MMDiff + DINO-v3 · PASCAL VOC 2012mIoU: 78.21 -> 84.95Improved
choose concept set object, background, detail, edges instead of no concepts (single-stream t=0) (ablation)MMDiff · PASCAL VOC 2012mIoU: 50.2 -> 63.8Improved
use a larger concept set with 23 class names plus generic concepts (ablation)MMDiff · PASCAL VOC 2012mIoU: 63.3 vs 63.8No clear effect
replace denoised latent with aggregated MMDiff featuresLyra video-to-3DGS decoder · RealEstate10KPSNR: 21.37 -> 21.61Improved
train a unified decoder jointly on VOC, DUTS and NYU (ablation)MMDiff · NYU Depth v2AbsRel: 0.116 -> 0.220Hurt

Later papers that compared with this method

1 later accepted paper reports MMDIFF as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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