Meet the Fellow: Shauli Ravfogel

NYU Center for Data Science
2 min readJul 24, 2024

--

This entry is part of our Meet the Fellow blog series, which introduces and highlights Faculty Fellows who have recently joined CDS.

Meet CDS Faculty Fellow Shauli Ravfogel, who will be joining us this fall. Ravfogel is currently completing his PhD in the Natural Language Processing Lab at Bar-Ilan University, supervised by Prof. Yoav Goldberg. He brings a wealth of experience in natural language processing, representation learning, and the analysis and interpretability of neural models.

Ravfogel’s research focuses on understanding how neural networks learn distributed representations that encode structured information, and how these representations are utilized to solve complex tasks. Ravfogel is particularly interested in developing techniques to control the content of these representations and map them back to interpretable concepts.

“I’m excited to join CDS and work at the intersection of various disciplines studying language,” said Ravfogel. “I look forward to collaborating with researchers from diverse backgrounds, including NLP, machine learning, and psycholinguistics, to gain deeper insights into the emergence of rich representations in language models.”

At CDS, Ravfogel plans to expand on his PhD work, which centered on controlling the content of neural representations. He aims to better understand the training dynamics that lead to the emergence of these rich representations and explore techniques like causal inference to subject language models to rigorous scientific inquiry.

Ravfogel’s research has been published in top-tier venues in the field of natural language processing. His work on linear encodings of human-interpretable concepts in language models, as presented in papers such as “Null It Out” and “Linear Adversarial Concept Erasure,” has opened new avenues for controlling and interpreting the information encoded in neural representations.

Prior to joining NYU, Ravfogel gained valuable industry experience through internships at Google Research Israel and the Allen Institute for AI. He has also served as a Machine Learning Lecturer at Bar-Ilan University, teaching an introduction to machine learning class for neuroscience students.

Ravfogel holds a BSc in both Computer Science and Chemistry from Bar-Ilan University, as well as an MSc in Computer Science from the same institution. This multidisciplinary foundation has informed his unique approach to computational linguistics and machine learning.

To view all our current faculty fellows, please visit the CDS Faculty Fellow page on our website.

By Stephen Thomas

--

--

NYU Center for Data Science

Official account of the Center for Data Science at NYU, home of the Undergraduate, Master’s, and Ph.D. programs in Data Science.