Update README.md

README.md

@@ -14,15 +14,42 @@ pipeline_tag: text-generation
 
 # SMB-v1-1.7B-Structure
 
-SMB-v1 Model Family is composed of biomedical multi-modal models for standard patient representations. We will release both the `SMB-v1-1.7B-Structure` (EHR Only) and the `SMB-v1-1.7B` (a complete multimodal version).
+## Model Type
+Multimodal Longitudinal Oncology Foundation Model
+
+## Model Description
+The SMB-v1-1.7B-Structure is the initial release of the SMB-v1 family, specifically engineered to model the complex, time-varying dynamics of cancer biology through structured clinical signals. It treats the clinical structured data as a multimodal environment, fusing heterogeneous data streams into a unified patient state representation.
+
+It is designed to ingest and synthesize diverse structured modalities across the patient journey, including:
+
+- Temporal Physiological Signals: Modeling continuous longitudinal trajectories of laboratory values, vital signs, and functional status markers to capture disease progression and physiological drift over time.
+
+- Clinical Events & Phenotypes: Encoding discrete, high-cardinality sequences of diagnosis codes (ICD), procedure events (CPT), and adverse events to reconstruct the semantic history of the patient's care.
+
+- Therapeutic Interventions: integrating complex treatment histories, including systemic therapies (chemotherapy, immunotherapy), radiation dosing schedules, and surgical interventions to understand causal treatment-response dynamics.
+
+- Molecular & Genomic Profiles: Embedding high-dimensional static and dynamic biomarker panels, including somatic mutations, gene expression signatures, and proteomic markers, directly alongside clinical phenotypes.
+
+- Oncologic Staging & Outcomes: Processing structured tumor staging (TNM), histology classifications, and survival endpoints to anchor representations in ground-truth biological states.
+
+## Intended Use Cases
+This model is optimized for downstream tasks requiring a deep understanding of longitudinal patient history, such as:
+
+- Predictive Risk Stratification: Forecasting adverse events, toxicity, or rapid progression based on historical trajectories.
+
+- Treatment Response Modeling: Simulating potential patient outcomes under different therapeutic regimens.
+
+- Patient Similarity Search: Identifying cohorts with similar biological and clinical progressions for real-world evidence generation.
+
+- Clinical Trial Matching: Aligning complex patient states with structured eligibility criteria.
+
+Note: While the full `SMB-v1-1.7B` will introduce unstructured modalities, this -Structure variant establishes the foundation using the highest-fidelity structured signals available in modern oncology data warehouses.
 
 ## Model Details
 
 - **LLM Backbone**: Qwen3-1.7B
-- **Vision Encoder**: None (text-only)
 - **Connector**: identity
 - **Model Family**: SMB-v1
-- **Modalities**: EHR Only
 - **Training Method**: SFT + JEPA Multi-objective
 
 ## Special Tokens