AI Paper Atlas

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

Circuit Mechanisms for Spatial Relation Generation in Diffusion Transformers

Binxu Wang, Jingxuan Fan, Xu Pan

CVPR 2026 · Image generation › Text-to-image diffusion · approach: Text-driven synthesis and semantic guidance · 3 citations (OpenAlex)

This paper tried a common change: Add positional encoding

add positional encoding to random token encoder · Synthetic relations · binding and relation accuracy Improved

What 646 accepted papers report for this change

See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.

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
ablate relation cross-attention head L2H8 (ablation)RTE-DiT-B · Synthetic relationsrelation accuracy: 67% -> 33% when removedImproved
ablate object head L4H3 (ablation)RTE-DiT-B · Synthetic relationsshape accuracy: 90% -> 76% when removedImproved
mask relation words in prompt (ablation)T5-DiT-B · Synthetic relationsrelation accuracyNo clear effect
mask shape2 text token (ablation)T5-DiT-B · Synthetic relationsshape/binding/relation accuracy: about 50% drop when maskedImproved
add positional encoding to random token encoder (ablation)RTE-DiT-B · Synthetic relationsbinding and relation accuracyImproved
use smallest model (DiT-nano) with RTE (ablation)RTE-DiT-nano · Synthetic relationsspatial relation accuracy: 5%Hurt
replace random token embeddings with T5 embeddingsDiT-B · Synthetic relationsin-distribution relation accuracyNo clear effect
insert filler word or reverse sentence in promptT5-DiT vs RTE-DiT · Synthetic relationsrelation accuracyHurt
vector arithmetic on shape2 T5 embedding to swap relation vector (ablation)T5-DiT · Synthetic relationsgenerated object positionImproved

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