AI Paper Atlas

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

Property-Informed Diffusion-Based Text-to-Microstructure Generation

Bingxuan Dai, Hongsong Wang, Jie Gui

CVPR 2026 · Image generation › Text-to-image diffusion · approach: Text-driven synthesis and semantic 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 property-informed conditioning (ablation)PropDiff-TMG · Geometries 2000Chamfer distance: 0.0651 -> 0.0395Improved
add property-informed conditioning (ablation)PropDiff-TMG · Geometries 2000FID: 105.35 -> 70.81Improved
add contrastive text-structure alignment (ablation)PropDiff-TMG · Geometries 2000FID: 264.63 -> 70.81Improved
add contrastive text-structure alignment (ablation)PropDiff-TMG · Geometries 2000CLIP score: 0.5161 -> 0.6936Improved
add test-time reward-guided alignment (ablation)PropDiff-TMG · Geometries 2000FID: 81.68 -> 70.81Improved
add test-time reward-guided alignment (ablation)PropDiff-TMG · Geometries 2000CLIP score: 0.6078 -> 0.6936Improved
add discriminator reward to text-alignment reward (ablation)PropDiff-TMG · Geometries 2000FID: 73.51 -> 70.81Improved
add discriminator reward to text-alignment reward (ablation)PropDiff-TMG · Geometries 2000CLIP score: 0.7038 -> 0.6936Hurt
add reward normalization (ablation)PropDiff-TMG · Geometries 2000CLIP score: 0.7189 -> 0.6936Hurt
add reward normalization (ablation)PropDiff-TMG · Geometries 2000FID: 77.52 -> 70.81Improved
add property-informed conditioning (ablation)PropDiff-TMG · GenText-MicrostructCLIP score: 0.5210 -> 0.6463Improved
add test-time reward-guided alignment (ablation)PropDiff-TMG · GenText-MicrostructCLIP score: 0.5164 -> 0.6463Improved

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