Shauli Ravfogel is a Faculty Fellow at the NYU Center for Data Science, specializing in the interpretability of large language models. With a PhD from Bar-Ilan University and a BSc in Chemistry, his research focuses on analyzing and controlling the internal representations of generative models.
He is a recipient of the Blavatnik Prize for Outstanding Israeli Doctoral Students in Computer Science.
Shauli Ravfogel
Postdoctoral Faculty Fellow at NYU Center for Data Science
Israel
Overview
Topics They Care About
AI Interpretability
His core research focuses on developing mathematically-principled approaches to understand and control the internal representations of large language models.
Causality in AI
He explores framing language models as causal models and is teaching a course on the causality and interpretability of LMs at NYU.
Mitigating AI Bias
A primary application of his work is developing methods, such as Iterative Nullspace Projection, to remove concepts like gender bias from neural representations.
Media Appearances
Shauli has no verified media appearances
Work History
1-2025
Postdoctoral Faculty Fellow at NYU Center for Data Science
10-2023 - 9-2024
Research Intern at Google
1-2020 - 6-2023
Research Intern at Allen Institute for AI (AI2)
10-2021 - 4-2022
Machine Learning Lecturer at Bar-Ilan University
2-2019
Machine Learning Teaching Assistant at Bar-Ilan University
Education
10-2020 - 4-2024
Doctor of Philosophy - PhD from Bar-Ilan University
2018 - 2020
Master's degree from Bar-Ilan University
More Information
Social Presence :
Prographics :
Exp : 4
Location : Israel
Job Level : Mid-senior
Designation : Postdoctoral Faculty Fellow at NYU Center for Data Science
