parameter-inference

A research paper presents a latent neural differential equation framework that infers unknown blood-clotting parameters from sparse measurements and forecasts thrombus growth, with stochastic neural ODEs achieving the best predictive performance.

This paper presents AgentODE, a framework that uses an LLM to propose ODE structures and a tool-augmented agent to refine parameter distributions from aggregate summary statistics alone, enabling mechanistic modeling of rare diseases under data scarcity and privacy constraints.