Hi! My name is Xiulin Yang and I’m currently a third-year PhD student in Computational Linguistics at Georgetown University, advised by Professor Nathan Schneider and Professor Ethan Wilcox. At Georgetown, I also work closely with Professor Amir Zeldes. I’m a member of NERT and PICoL.

Before that, I have been lucky to work with Professor Jamie Findlay and Professor Hanne Eckhoff at Oxford, Professor Alexander Koller at Saarland and Professor Johan Bos at Groningen.

I am broadly interested in how language is learned and generalized by humans and machines. Specifically, my research explores three related questions:

• What makes language learnable? Are some languages harder for machines to learn than others, and if so, what causes these differences? How can we bridge these gaps across languages? (Yang et al., 2026; Yang et al., 2025)
• Whether and how do language models generalize beyond their training experience? (Yang, 2025; Yang et al., 2023)
• What inductive biases enable humans to learn language efficiently, and can these biases improve language models? (Yang et al., 2026)

I approach these questions through work on multilingual/monolingual/crosslingual language learning, tokenization, targeted evaluation, interpretability, and cognitively inspired NLP.

When I’m not in front of the screen, you can find me bouldering/wall-climbing, reading (I’m a big fan of science fictions!), or swimming.

📰 News

Last updated: August 2026