I’m a PhD student studying Linguistics at Harvard University, with a focus on computational linguistics and semantics.
I’m interested in how we can use linguistics and experimental data to pin down language models’ capabilities and limitations and develop ways to improve them. For all their impressive performance, we’re still far from understanding exactly how language models achieve it – and why they still fail to handle phenomena from simple negation to entity tracking. I’m currently investigating whether large language models can compose meanings into larger pieces, specifically whether they generalize to new cases of adjective-noun composition.
In additional to language models, I also work on experimental semantics with Kathryn Davidson and Gennaro Chierchia. We’re developing a robust experiment design to test pronoun reference in crossover configurations, which quickly become difficult to judge outside of textbook cases.