Hello! I’m a postdoctoral fellow in the Department of Computer Science at Boston College, advised by Emily Tucker Prud’hommeaux. I received my PhD in Linguistics from the University of California, Davis in Summer 2020, advised by Kenji Sagae.
My research studies structural variation in both languages and machines. I address related questions from the angle of language typology, low-resource NLP, and cognitive development, along with a non-western mind. The general methodology that I take upon is a data-driven approach coupled with The #BenderRule and methods of number counting at varying degrees of carbon dioxide consumption.
Besides academic responsibilities, I proudly serve on the planning committee for Advocates for Indigenous California Language Survival, and their biennial institute Breath of Life.
My non-research interests are music, food, and simple methods, and in that order too.
Fall 2020 -
Started working as a postdoc in late November. I shift focus from analyzing large-scale corpora to low-resource settings, splitting time between endangered language documentation and language development. For the former, I focus on capturing structural variation via development of language generation technologies for the Iroquoian language family in order to help the communities with their own language reclamation. Most of the Iroquoian languages have very few or no living native/fluent speakers whose communities are currently engaged in efforts to reclaim their heritage languages. For the latter, I focus on capturing syntactic and pragmatic variation comparing children and adults with different neurodevelopmental conditions, using computational and experimental methods.
2014 - Summer 2020
During this time, I got a PhD. My dissertation project focuses on crosslinguistic modeling of word order preferences, asking what abstract constraints as well as idiosyncrasies govern language users’ structural choices from both synchronic and diachronic perspectives. Additionally, this project also tries to adapt theoretical framework of dependency syntax to develop a dependency treebank for Hupa, an endangered Dene language of northwestern California traditionally spoken in Hoopa Valley on the lower Trinity River in present-day Humboldt County, as a way to formalize and model the syntax of indigenous languages.
During this time, I interned as a software developer at the Cognitive Computing Lab at Baidu. I worked on designing graph represetations for open-domain information extraction in English and Mandarin.