Hi! I’m Aryaman, an incoming Ph.D. student in Computer Science in the Stanford NLP group. Currently, I am a final-year undergrad majoring in Computer Science and Linguistics at Georgetown University, where I am advised by Nathan Schneider. I also frequently collaborate with Ryan Cotterell at ETH Zürich.
My research interests are multilingual NLP, computational linguistics, and mechanistic interpretability. I also enjoy historical linguistics. Some fundamental questions I care about are:
- How do transformer language models work? Why are they so successful and alternatives (like RNNs) are not?
- How do multilingual language models distribute and share processing across languages?
- Can we formalise some ideas about human language from linguistics using information-theoretic methods?
- What is the linguistic history of the Indian subcontinent?
You’ll find more on my résumé and list of publications. I also have some small NLP-focused blog posts here.
- 2023-05-24: Two workshop papers accepted:
- Jambu: A historical linguistic database for South Asian languages (SIGMORPHON)
- Unified syntactic annotation of English in the CGEL framework (LAW)
- 2023-02-08: Accepted to the Ph.D. program at Stanford CS!
- 2022-12-18: Headed to Berkeley, CA for a 5-week internship at Redwood Research on mechanistic interpretability.
# Selected papers
- Localizing model behavior with path patching
arXiv, arXiv:2304.05969 • 2023 Apr 12
- Estimating the entropy of linguistic distributions
ACL • 2022 Apr 04