Header

Jason mai will introduce the basics of technical obstacles and inequalities that LLMs face when shifting towards a multilingual model with focus on monolingual bias that LLM’s face as they are trained with English and the Language Tokenization costs with perfmance benchmarks.

Mani Deepika will introduce how LLM’s help in transciption by annotating the morphosyntactic Analysis pipeline thereby reducing the manual annotation process. The talk also focuses on the Automatic Morphosyntactic tagging along with the limitations that we are encountering and the future directions.