In the news
Target leakage, not model class, explains reported accuracy in survey-based cardiovascular screening: a leakage-tiered audit of glass-box and tabular foundation models
arXiv cs.AI · Published · 3 min read
In 30 seconds
- What happened
- Researchers audited cardiovascular screening models and found that reported high accuracy comes from data leakage in features, not model sophistication.
- Why it matters
- Healthcare engineers building screening systems should care when evaluating model performance claims from survey-based health datasets.
- Watch out
- The study used specific datasets and thresholds; results may not generalize to other screening tasks or populations with different demographic distributions.
- foundation model
- eval
- benchmark
The patterns behind this
Each one covers how the technique works, when it earns its cost, and where it breaks.
The Agent Architect
One pattern, one tradeoff, one production failure story. A short weekly briefing for people building agentic systems.
Weekly email, one-click unsubscribe. We only use your address to send the briefing.