Giulio Franzese

Discrete diffusion directly estimates information-theoretic quantities, outperforming continuous embedding methods on real-world data without complex engineering.

DIPHINE reveals the intricate information dynamics of complex systems, outperforming traditional methods and enabling insights from real-world data without distributional constraints.

Current evaluation metrics for trajectory inference can mislead researchers, but functional KL divergence offers a clearer, more reliable comparison of methods in sparse data conditions.