AI Paper Atlas

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

Parallel Multimodal Diffusion Language Models for Thinking-Aware Editing and Generation

Ye Tian, Ling Yang, JiongFan Yang, Anran Wang, Yu Tian, Jiani zheng, Haochen Wang, Zhiyang Teng, Zhuochen Wang, Yinjie Wang, Yunhai Tong, Mengdi Wang and 1 more

ICLR 2026 · Image generation › Text-to-image diffusion · approach: Compositional text-to-image guidance

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
use parallel instead of sequential text-then-image denoising (ablation)MMaDA-Parallel · ParaBenchOutput Alignment: 48.9 -> 51.5Improved
use parallel instead of sequential decoding (ablation)MMaDA-Parallel · ParaBenchText Alignment: 70.6 -> 70.4No clear effect
add ParaRL post-training (ablation)MMaDA-Parallel · ParaBenchOutput Alignment: 51.5 -> 59.8Improved
use trajectory-level rewards instead of output-level RL (ablation)ParaRL · ParaBenchOutput Alignment: 53.6 -> 59.8Improved
use output-level RL over no RL (ablation)MMaDA-Parallel · ParaBenchOutput Alignment: 51.5 -> 53.6Improved
increase ParaRL sampled steps from 2 to 3 (ablation)ParaRL · ParaBenchOutput Alignment: 53.6 -> 59.8Improved
increase ParaRL sampled steps from 3 to 4 (ablation)ParaRL · ParaBenchOutput Alignment: 59.8 -> 58.7No clear effect
enable thinking in Bagel on spatial editsBagel · ParaBench editing (Spatial)editing score: 69.8 -> 65.0Hurt
enable thinking in Bagel on temporal editsBagel · ParaBench editing (Temporal)editing score: 72.3 -> 75.6Improved
apply parallel framework and ParaRL to Lumina-DiMOO (ablation)Lumina-DiMOO · ParaBenchOutput Alignment: 63.3 -> 68.8Improved
use MMaDA-Parallel instead of sequential MMaDA on GenEvalMMaDA · GenEvalOverall: 0.68 -> 0.71Improved
use MMaDA-Parallel instead of sequential MMaDA on RISEBenchMMaDA · RISEBenchOverall accuracy: 5.5% -> 5.75%Improved

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