Kayo Yin

UC Berkeley

Kayo Yin is a first-year PhD student at UC Berkeley advised by Jacob Steinhardt and Dan Klein. She currently works on interpretability, sign language processing, and AI for music. Before that, she was a Master’s student at Carnegie Mellon University advised by Graham Neubig, and she completed her undergraduate studies at École Polytechnique in 2020. Her research has been recognized by an EMNLP Best Paper Honorable Mention award, an ACL Best Theme Paper award, and the Thomas Clarkson medal, and she is the recipient of a Siebel Scholarship. She also really likes playing music, snowboarding, and exploring the outdoors.

Natural Language Processing for Signed Languages

Signed languages are the primary means of communication for many deaf and hard-of-hearing individuals. Since signed languages exhibit all the fundamental linguistic properties of natural language, we believe that tools and theories of Natural Language Processing (NLP) are crucial for its modeling. However, existing research in Sign Language Processing (SLP) seldom attempts to explore and leverage the linguistic organization of signed languages. In this talk, I will talk about why I believe NLP researchers should include signed languages in their research and some of the challenges in sign language processing today. I will also cover two projects that take steps in extending NLP to signed languages: data augmentation for sign language translation and coreference resolution for pronominal indexing signs. Through these examples, I will illustrate that NLP plays a crucial role in the development of sign language processing systems.