# Text-to-Image Synthesis

4 related articles

Subscribe to updates

Scientific Reports Nov 10, 2026
Otoendoscopic images stop relying on text alone: CLNA turns unwritten visual variation into an adjustable latent, cutting FID from 52.91 to 10.11
Otoendoscopic image synthesis has been limited because text prompts state a diagnosis but not visual details such as hue and texture, leaving appearance variation within a diagnostic class hard to control. CLNA adds a style encoder and dual cross-attention to a Stable Diffusion 2.1 backbone, compressing such non-prompt attributes into a continuous latent whose sampling range is set by a style parameter. Across normal tympanic membrane, otitis media with effusion and chronic otitis media, FID against the held-out internal test set fell from 52.91 for baseline Stable Diffusion to 10.11, and feature-space recall rose from 57.67% to 90.22%.
Otoendoscopic image synthesis has been limited because text prompts state a diagnosis but not visual details such as hue and texture, leaving appearance variation within a diagnostic class hard to control. CLNA adds a style encoder and dual cross-attention to a Stable Diffusion 2.1 backbone, compressing such non-prompt attributes into a continuous latent whose sampling range is set by a style parameter. Across normal tympanic membrane, otitis media with effusion and chronic otitis media, FID against the held-out internal test set fell from 52.91 for baseline Stable Diffusion to 10.11, and feature-space recall rose from 57.67% to 90.22%.
University of Ulsan
Keunwoo Park（First author）Joong Ho Ahn（Corresponding author）Jihoon Kweon（Corresponding author）
CLNA Text-to-Image Synthesis Stable Diffusion Data Augmentation
arXiv Oct 14, 2026
Pathology generation leaves the 1024-pixel tile behind: coherent gigapixel slides, and synthetic samples lift few-shot classification by up to 18%
Histopathology image generation has stayed at the patch scale, producing tiles of at most 1024×1024 pixels, while clinical text reports and slide-level diagnostic labels act on the whole slide. BigCell trains a flow-matching transformer at 2048×2048 pixels on over 120,000 public whole-slide images and uses an inference-time sliding-window pipeline to generate coherent images up to 32,768×32,768 pixels, outperforming prior tile generators on tile-level and slide-level fidelity metrics. An adaptive scheduling strategy speeds generation by over three times with minimal quality degradation, and synthetic samples improve slide-level few-shot classification by up to 18%.
Histopathology image generation has stayed at the patch scale, producing tiles of at most 1024×1024 pixels, while clinical text reports and slide-level diagnostic labels act on the whole slide. BigCell trains a flow-matching transformer at 2048×2048 pixels on over 120,000 public whole-slide images and uses an inference-time sliding-window pipeline to generate coherent images up to 32,768×32,768 pixels, outperforming prior tile generators on tile-level and slide-level fidelity metrics. An adaptive scheduling strategy speeds generation by over three times with minimal quality degradation, and synthetic samples improve slide-level few-shot classification by up to 18%.
Srikar Yellapragada（First author）
BigCell Flow-Matching Transformer Text-to-Image Synthesis
arXiv Oct 2, 2026
PixelDense splits SAM2, Depth Anything v2 and Metric3D v2 into semantic and geometric alignment streams, lifting PixelGen-XXL's GenEval Overall from 0.7927 to 0.8093
The work introduces PixelDense, which routes DINOv2 and SAM2 through a semantic projection stream and Depth Anything v2 and Metric3D v2 through a geometric projection stream, adding a weight-space orthogonality penalty that keeps the two streams in disjoint subspaces; with a single recipe on PixelGen and DeCo it improves GenEval, DPG-Bench and HPS v2.1, raising PixelGen-XXL's GenEval Overall from 0.7927 to 0.8093 and beating every single-teacher and unfactored multi-teacher variant.
The work introduces PixelDense, which routes DINOv2 and SAM2 through a semantic projection stream and Depth Anything v2 and Metric3D v2 through a geometric projection stream, adding a weight-space orthogonality penalty that keeps the two streams in disjoint subspaces; with a single recipe on PixelGen and DeCo it improves GenEval, DPG-Bench and HPS v2.1, raising PixelGen-XXL's GenEval Overall from 0.7927 to 0.8093 and beating every single-teacher and unfactored multi-teacher variant.
Lehan Yang（First author）
PixelDense Visual Representation Alignment Pixel-Space Diffusion Transformer Text-to-Image Synthesis Image Editing
arXiv Sep 30, 2026
Adversarial post-training restores missing high-frequency detail in pixel diffusion, cutting DeCo FID from 33.27 to 28.59
By keeping the original diffusion or flow-matching objective and adding an adversarial loss to the predicted output at non-high-noise timesteps, this study jointly improves distribution fidelity, coverage, prompt alignment, and perceptual quality on the DeCo and PixelGen pixel backbones, attributing the gain to restored missing natural-image high-frequency power, while the same procedure yields no comparable gain in the tested latent diffusion configurations.
By keeping the original diffusion or flow-matching objective and adding an adversarial loss to the predicted output at non-high-noise timesteps, this study jointly improves distribution fidelity, coverage, prompt alignment, and perceptual quality on the DeCo and PixelGen pixel backbones, attributing the gain to restored missing natural-image high-frequency power, while the same procedure yields no comparable gain in the tested latent diffusion configurations.
Xin Lin（First author）
adversarial training Pixel Diffusion Text-to-Image Synthesis

You’re all caught up

Otoendoscopic images stop relying on text alone: CLNA turns unwritten visual variation into an adjustable latent, cutting FID from 52.91 to 10.11

Pathology generation leaves the 1024-pixel tile behind: coherent gigapixel slides, and synthetic samples lift few-shot classification by up to 18%

PixelDense splits SAM2, Depth Anything v2 and Metric3D v2 into semantic and geometric alignment streams, lifting PixelGen-XXL's GenEval Overall from 0.7927 to 0.8093

Adversarial post-training restores missing high-frequency detail in pixel diffusion, cutting DeCo FID from 33.27 to 28.59