I’m a 1st year PhD student in NLP/ML at University of Southern California, working with Dani Yogatama. I’m interested in interpretability - how we make sense of machine learning models, and how interpretability might uncover the science underpinning large-scale models. I’m particularly interested in Emergence/Self-Organization, which I loosely describe as component-level interactions leading to a system-wide pattern. My goal is to study phenomena like that with tools like interpretability. I used to study Physics and Complex Systems, so I borrow some approaches from that toolbox. I also study emergent abilities of large models such as reasoning.

Besides research, I read/write, run, and I have a love-hate relationship with nature. e.g. I like large bodies of water - one time I threw my phone at a sea gull (in self defense).