I am a fourth year PhD student in the department of statistics at the University of Michigan, where I am advised by Prof. Yuekai Sun and Prof. Ya’acov Ritov. Prior to attending the University of Michigan, I recieved my bachelors degree in mathematics from Colorado State University, where I worked with Prof. Renzo Cavalieri on probems in algebraic geometry. During this time I was supported by the Barry Goldwater Scholarship.

Research Highlights

Broadly, my research interests are on the statistical aspects of making AI safe, fair, and more efficient.

On the fairness side, my research has pertained to leveraging the effects of distribution shift to achieve greater group fairness guarantees [1], while my work on AI safety includes introducing a transfer learning framework and label refinement method for the weak to strong generalization problem [2].

Recently, I worked with collaborators at the University of Waterloo and IBM research to release a new LLM router, CARROT [3]. For a a more extensive list of publications, see the research tab.

[1] Algorithmic Fairness in Performative Policy Learning: Escaping the Impossibility of Group Fairness. S Somerstep, Y Ritov, Y Sun. FAccT 2024

[2] A transfer learning framework for weak to strong generalization S Somerstep, F Polo, M Banerjee, Y Ritov, M Yurochkin, Y Sun. ICLR 2025

[3] CARROT: A Cost Aware Rate Optimal Router. S Somerstep, F Polo, A Oliveira, P Mangal, M Silva, O Bhardwaj, M Yurochkin, S Maity. Under Review