/ælɪks wɔɹstæt/ 🔊
I’m a fifth year PhD student in linguistics at New York University working with Sam Bowman.
I study the learnability of grammar using artificial neural networks. In particular, I’ve been investigating what the success of deep learning can teach us about the human language acquisition and the poverty of the stimulus.
I have created several challenge datasets for neural networks incorporating insights from theoretical linguistics. These include the CoLA and BLiMP datasets of acceptability judgments, the NOPE and ImpPres datasets for evaluating pragmatic knowledge of natural language inference (NLI) models. I also do work in formal and experimental semantics and pragmatics. I’m interested in theories of relevance and the role of questions in discourse.
alexwarstadt <at> gmail <dotcom>
warstadt <at> nyu <dotedu>