Florentina Bunea is currently at Cornell University, the Department of Statistics and Data Science in the Bowers College of Computing and Information Science. Her research interests lie at the intersection of mathematical statistics and interpretable statistical machine learning.

A major focus of her recent work is in high dimensional inference for interpretable models with hidden structure. The research agenda includes the provision of minimal constructive identifiability assumptions for latent factor models, the derivation of statistical (minimax ) limits of estimation in these models and of computationally feasible, (near) minimax rate adaptive, algorithms; the derivation of distributional limits of estimators associated with the hidden structure; prediction with hidden structures (interpolation versus interpretation). Latent space overlapping clustering, (sparse) topic models and their comparison in terms of the Wasserstein distance provide motivating applications of this general theme.