AI Paper Atlas

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

Revisiting Global Text Conditioning in Diffusion Transformers

Nikita Starodubcev, Daniil Pakhomov, Zongze Wu, Ilya Drobyshevskiy, Yuchen Liu, Zhonghao Wang, Yuqian Zhou, Zhe Lin, Dmitry Baranchuk

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
remove pooled CLIP embedding (set to zero) for long prompts (ablation)FLUX schnell · MJHQ long promptsImageReward: 10.3 -> 10.4No clear effect
remove pooled CLIP embedding for short prompts (ablation)FLUX schnell · MJHQ short promptsImageReward: 6.2 -> 4.5Improved
remove pooled CLIP embedding (ablation)HiDream-Fast · MJHQ short and long promptsImageReward: 7.9 -> 8.1 (short), 12.8 -> 13.0 (long)No clear effect
add pooled CLIP embedding to CLIP-free COSMOS without guidance (ablation)COSMOS · Side-by-side complexity win ratecomplexity win rate (%): 43Hurt
add modulation guidance (aesthetics)FLUX schnell · COCO 5kImageReward: 10.2 -> 11.0Improved
add modulation guidanceFLUX schnell · GenEval object countingscore: 56 -> 65Improved
add modulation guidance on CausVidCausVid 1.3B · VBenchdynamic degree: 75.25 -> 86.59Improved
add pooled CLIP embedding without guidance (ablation)CausVid 1.3B · VBenchtotal score: 62.72 -> 62.82No clear effect
use dynamic layer-wise guidance instead of constant scale (ablation)modulation guidance on FLUX schnell · hands correction, human SbSwin rate (%): 48 -> 59Improved
use dynamic step guidance instead of constant scale (ablation)modulation guidance on FLUX schnell · aesthetics, human SbSwin rate (%): 62 -> 72Improved
use modulation guidance instead of Normalized Attention GuidanceFLUX schnell · aesthetics, human SbSwin rate (%): 33 -> 67Improved
use modulation guidance instead of Concept SlidersFLUX schnell · hands defects, human SbSwin rate (%): 42 -> 58Improved

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