Prior-Fitted Functional Flow: In-Context Generative Models for Pharmacokinetics
It provides a novel generative modeling approach for pharmacokinetics, enabling automated and accurate population-level and individual-level predictions, which is important for drug development and personalized medicine.
This paper introduces Prior-Fitted Functional Flows, a generative foundation model for pharmacokinetics that achieves zero-shot population synthesis and individual forecasting without manual tuning, demonstrating state-of-the-art predictive accuracy on real-world datasets.
We introduce Prior-Fitted Functional Flows, a generative foundation model for pharmacokinetics that enables zero-shot population synthesis and individual forecasting without manual parameter tuning. We learn functional vector fields, explicitly conditioned on the sparse, irregular data of an entire study population. This enables the generation of coherent virtual cohorts as well as forecasting of partially observed patient trajectories with calibrated uncertainty. We construct a new open-access literature corpus to inform our priors, and demonstrate state-of-the-art predictive accuracy on extensive real-world datasets.