Events

Colloquium

Statistical Attribute Alignment for Black-Box Generative AI

Speaker: Jason Klusowski (Princeton) Associate Professor in the Department of Operations Research and Financial Engineering (ORFE) Princeton University Wednesday, October 7, 2026 12:00PM - 1:00PM Lunch 11:30am-12:00pm Talk 12:00pm-1:00pm Location: Yale Institute for Foundations of Data Science & Webcast, 219 Prospect Street, 13th Floor, New Haven, CT 06511 and via Webcast: https://yale.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=235c1f91-416f-4e4f-be13-b4d9011dd0e2

Abstract: Generative AI offers a powerful way to create synthetic data for simulation, training, and evaluation. But what if a model generates outputs with an attribute distribution that differs from what a downstream task requires? For example, medical research may require image collections with specified distributions of demographic groups or disease categories. This talk develops a statistical framework for attribute alignment through output post-processing, using only query access to the generative model. After observing the attributes of generated outputs, we select a batch whose attributes jointly follow a target distribution, exactly or approximately. We develop sampling algorithms and establish their asymptotic optimality in expected model query cost as the desired batch size grows.

Speaker Bio: Jason M. Klusowski is an Associate Professor in the Department of Operations Research and Financial Engineering (ORFE) at Princeton University. He is also affiliated with the Princeton Laboratory for Artificial Intelligence (AI Lab) and the Princeton Language and Intelligence (PLI) initiative. He studies the mathematical and statistical foundations of machine learning and AI systems, examining how data, computation, and model structure shape their capabilities and limitations.

Before joining Princeton, Jason was an Assistant Professor in the Department of Statistics at Rutgers University–New Brunswick. He received his Ph.D. in Statistics and Data Science from Yale University.

Jason serves on the editorial board of Bernoulli, the journal of the Bernoulli Society. His research is partially supported by a Sloan Research Fellowship in Mathematics and NSF CAREER DMS-2239448, and was previously supported by NSF DMS-2054808 and TRIPODS DATA-INSPIRE Institute CCF-1934924.

Jason grew up in Winnipeg, in the heart of the Canadian Prairies. His spouse is an Assistant Professor of Marketing at Yale University.

Add To: Google Calendar | Outlook | iCal File

• Colloquium