Papers by Bill Dyer

1 papers
Leveraging Syntactic Dependencies in Disambiguation: The Case of African American English (2024.lrec-main)

Copied to clipboard

Challenge: African American English (AAE) is a low-resource language facing the challenge of inadequate annotated data for training natural language processing models.
Approach: They propose a syntactically informed classifier for automatic disambiguation of AAE's habitual be.
Outcome: The proposed classifier improves automatic disambiguation of habitual and non-habitual meanings of "be" integrating syntactic information improves disambiguations of habituality by 65 F1 points over baseline models and as much as 74 points.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations