I’m a sixth year linguistics PhD student at New York University working with Tal Linzen in the Computation and Psycholinguistics Lab. My current work explores what training setups produce models that remember, forget, and teach the way people do in interaction. I study this question using behavioral experiments with human participants, model evaluation, and post-training. Previously, I approached this question from the pre-training side, asking what the linguistic input must contain for a statistical learner to generalize the way people do.

Before coming to NYU, I completed my BA in linguistics and BS in computer science at the National University of Singapore advised by Michael Yoshitaka Erlewine and Min-Yen Kan.

Outside of linguistics, I enjoy ceramics, baking, and puppy pictures.

Recent Publications

2026

Manipulating language models’ training data to study syntactic constraint learning: The case of English passivization

Cara Su-Yi Leong and Tal Linzen

Journal of Memory and Language

Paper

2024

Prompting sometimes invokes expert-like downward shifts in GPT-4V’s conceptual hierarchies

Cara Su-Yi Leong and Brenden Lake

CogSci 2024

PaperPoster

2023

Language Models Can Learn Exceptions to Syntactic Rules

Cara Su-Yi Leong and Tal Linzen

SCiL 2023

PaperPoster