VGL-stGen: Learning a shared multimodal model for lung histology and spatial transcriptomics
VGL-stGen demonstrates how a medical vision-language backbone can learn from spatial transcriptomics and histology to connect tissue morphology and molecular states through shared generative modeling.