We prove clean optimality results for clustering in ultrametrics, identify ways to take advantage of this theory and thoroughly evaluate the resulting techniques.

LeapFactual is a model-agnostic algorithm that generates reliable, actionable counterfactual explanations using conditional flow matching, overcoming key limitations of existing methods in high-stakes and human-in-the-loop domains.

Predictive coding principles proceeding over modulated representations continually integrate top-down information from sparse labels in unsupervised representation learning.