SpaVis-6M (ICLR 2026)

Fusing Pixels and Genes: Spatially-Aware Learning in Computational Pathology

Paper | Code | Model | Dataset |

Abstract: Recent years have witnessed remarkable progress in multimodal learning within computational pathology. Existing models primarily rely on vision and language modalities; however, language alone lacks molecular specificity and offers limited pathological supervision, leading to representational bottlenecks. In this paper, we propose STAMP, a Spatial Transcriptomics-Augmented Multimodal Pathology representation learning framework that integrates spatially-resolved gene expression profiles to enable molecule-guided joint embedding of pathology images and transcriptomic data. Our study shows that self-supervised, gene-guided training provides a robust and task-agnostic signal for learning pathology image representations. Incorporating spatial context and multi-scale information further enhances model performance and generalizability. To support this, we constructed SpaVis-6M, the largest Visium-based spatial transcriptomics dataset to date, and trained a spatially-aware gene encoder on this resource. Leveraging hierarchical multi-scale contrastive alignment and cross-scale patch localization mechanisms, STAMP effectively aligns spatial transcriptomics with pathology images, capturing spatial structure and molecular variation. We validate STAMP across six datasets and four downstream tasks, where it consistently achieves strong performance. These results highlight the value and necessity of integrating spatially resolved molecular supervision for advancing multimodal learning in computational pathology.

---

Highlights

• Molecule-guided supervision: uses spatial transcriptomics gene expression as a task-agnostic supervisory signal for pathology representation learning.
• Spatial-aware gene encoder: models spatial co-localization patterns among Visium spots at scale (SpaVis-6M).
• Hierarchical multi-scale alignment: aligns pathology image patches and gene profiles with multi-scale contrastive objectives.
• Cross-scale localization: encourages patch-level spatial grounding across magnifications.

Data Overview

SpaVis-6M (dataset)

SpaVis-6M is a large-scale Visium-based spatial transcriptomics corpus used to train the spatial-aware gene encoder:

• 5.75M spatial transcriptomics entries
• 35 organs, 1,982 slices, 262 datasets/publications
• Alignment pretraining uses 697K paired pathology image–gene expression samples

Data note: We only provide processed GEO-derived data; for STimage-1K4M, HEST-1K, SpatialOmics, and STOmicsDB, please refer to their official websites.

Citation

If you find our work useful, please cite:

@article{STAMP_TBD,
  title   = {Fusing Pixels and Genes: Spatially-Aware Learning in Computational Pathology},
  author  = {TBD},
  journal = {TBD},
  year    = {TBD}
}