Yuwei Zhang

Inertial activity recognition models can now learn rich, global motion structure from unlabeled data by discovering and grouping motion primitives, achieving state-of-the-art performance with orders of magnitude less data.

Zero-shot LLM agents struggle to predict wellbeing scores from longitudinal data, often performing no better than a basic mean baseline.

Consumers are leveraging AI for financial insights but are hesitant to delegate execution, highlighting a critical gap in human-AI collaboration.

WEQA achieves a 24% accuracy boost in wearable health question answering by dynamically adapting to the complexities of sensor data and user queries.

Kairos achieves top-tier performance in Physical AI while ensuring efficient state management over extended time horizons, setting a new standard for operational world models.

Training a foundation model on a trillion minutes of wearable sensor data unlocks surprisingly accurate predictions across a wide range of health conditions, even with limited labeled data.