Aryaman Arora
I am a prospective Ph.D. student in NLP for Fall 2023. But besides that, I would say I am some combination of the following:
- Third-year undergraduate at Georgetown University in Washington, D.C. studying computer science and linguistics.
- Computational linguistics/NLP researcher advised by Dr. Nathan Schneider. My research interests are multilingual NLP for South Asia, computational linguistics (particularly through information theory), and historical linguistics. Lately I am also interested in mechanistic interpretability. See my publications
- Lover of languages: I speak Hindi–Urdu and English natively, and have studied Mandarin for some years. I have learned many more South Asian languages to varying degrees, including Punjabi (a heritage language of mine), Sindhi, Sanskrit, and so on. In language documentation, I have work on the Iranian minority language Kholosi.
- Dec 2022— Research Resident at Redwood Research trying to make the mechanisms of language models human-interpretable. More details coming soon.
- Appreciator of spicy foods from all cultures, Urdū šāʿirī, Aśokan dʰammalipi-s, segment trees (my favourite data structure), well-designed Unicode fonts, and websites written in raw HTML.
In the past, I was also some of these things:
- May—Aug 2022 AI/ML Intern at Apple, where I evaluated the robustness of Siri's natural language understanding systems. (Fun fact: I did not and do not own an iPhone.)
- June—Jul 2021 Research Intern at ETH Zürich in Switzerland with Dr. Ryan Cotterell. This is where my interest in information theory began and continues to grow.
You can see my resume here. Also contact me through Twitter.