A Multimodal Foundation Model for Longitudinal Patient Representation and Scalable Insight Generation in Oncology
cs.LG
Submitted: 2026-08-25
Updated: 2026-08-25
Terminology
Sources
- The Patient is not a Moving Document: A World Model Training Paradigm for Longitudinal EHR
- Sparse Autoencoders Find Highly Interpretable Features in Language Models
- The Faiss library
- CEHR-BERT: Incorporating temporal information from structured EHR data to improve prediction tasks
- MedGemma 1.5 Technical Report
- Language Models Are An Effective Patient Representation Learning Technique For Electronic Health Record Data
- MOTOR: A Time-To-Event Foundation Model For Structured Medical Records
- Molecular-driven Foundation Model for Oncologic Pathology
- Screen Them All: High-Throughput Pan-Cancer Genetic and Phenotypic Biomarker Screening from H&E Whole Slide Images
- Generative Medical Event Models Improve with Scale
- A multimodal and temporal foundation model for virtual patient representations at healthcare system scale
- Virchow2: Scaling Self-Supervised Mixed Magnification Models in Pathology
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks