Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
EHRSHOT: An EHR Benchmark for Few-Shot Evaluation of Foundation Models
Michael Wornow, Rahul Thapa, Ethan Steinberg, Jason Alan Fries, Nigam Shah
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use pretrained CLMBR-T-base (frozen, logistic head) instead of count-based GBM trained from scratch | CLMBR-T-base · EHRSHOT, 4 task categories aggregated | AUROC | Improved |
| pretrained CLMBR-T-base vs count GBM across all k | CLMBR-T-base · Operational Outcomes | AUROC | Improved |
| pretrained CLMBR-T-base vs count GBM at intermediate k (8 to 128) | CLMBR-T-base · Anticipating Lab Test Results and Chest X-ray Findings | AUROC | Improved |
| pretrained CLMBR-T-base vs count GBM at k = 1 | CLMBR-T-base · EHRSHOT | AUROC | No clear effect |
| increase number of labeled examples k toward full training set | CLMBR-T-base vs GBM · EHRSHOT | AUROC gap | Hurt |
| pretrained CLMBR-T-base vs count GBM at k > 64 | CLMBR-T-base · Assignment of New Diagnoses | AUROC | Hurt |
| Logistic Regression and Random Forest instead of LightGBM on count features | count-based baseline · EHRSHOT | AUROC | No clear effect |
Closest papers by meaning
- Context Clues: Evaluating Long Context Models for Clinical Prediction Tasks on EHR Data (ICLR 2025 · similarity 0.90)
- One Loss to Rule Them All: Marked Time-to-Event for Structured EHR Foundation Models (NeurIPS 2026 · similarity 0.87)
- Yet Another ICU Benchmark: A Flexible Multi-Center Framework for Clinical ML (ICLR 2024 · similarity 0.86)
- Contrastive Learning for Clinical Outcome Prediction with Partial Data Sources (ICML 2024 · similarity 0.86)
- ACES: Automatic Cohort Extraction System for Event-Stream Datasets (ICLR 2025 · similarity 0.85)
- GraphCare: Enhancing Healthcare Predictions with Personalized Knowledge Graphs (ICLR 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
