Dr. Di Liu will present an NRT LinDiv organized colloquium on Friday, February 16, 2024 at 9:00AM in the Foster Auditorium on PSU University Park campus and via Zoom.
Talk title TBA.
More information about Dr. Liu is available at the webpage below:
https://education.temple.edu/about/faculty-staff/di-liu-tum92163
The application for the NRT LINDIV graduate training program for the Fall 2024 cohort is open! Deadline to apply is February 28, 2024. Please attend one of the information sessions
To be eligible to apply for the graduate training program you must be a 1st or 2nd year doctoral student in the following programs:
The goal of the LinDiv graduate training program is to educate a new generation of experts in human-technology interaction who can bridge the gap between human linguistic diversity and technological linguistic illiteracy. NSF funding is limited to students holding U.S. citizenship or residency; a limited number of internally-funded fellowships are available for international students.
The deadline to apply for the NRT LINDIV graduate training program for the Fall 2023 cohort is fast approaching! To be eligible to apply for the graduate training program you must be a 1st or 2nd year doctoral student in the following programs:
The goal of the LinDiv graduate training program is to educate a new generation of experts in human-technology interaction who can bridge the gap between human linguistic diversity and technological linguistic illiteracy. NSF funding is limited to students holding U.S. citizenship or residency; a limited number of internally-funded fellowships are available for international students.
Friday, February 10, 2023, 9:00–10:30 a.m. EST, 127 Moore Building and virtually via Zoom
Dr. Shomir Wilson
Assistant Professor and Director of the Human
Language Technologies Lab in the
College of Information Sciences and Technology
at Penn State
“Sociodemographic Biases in Natural Language Processing: Two Case Studies”
Large language models (LLMs) are widely used in natural language processing (NLP) to
obtain high performance on a variety of tasks. However, the large corpora used to train
these models contain sociodemographic biases, and LLMs tend to inherit those biases, with
potentially harmful results. Shomir Wilson will present two case studies that reveal the
sociodemographic biases of select LLMs within the context of sentiment analysis, a
common NLP task. The first study shows that Word2Vec and GloVe exhibit negative
sentiment bias toward terms for people with disabilities. The second study shows that GPT-
2 exhibits a range of sentiment biases for nationality demonyms, i.e., words that specify
national origins. Shomir will conclude with some thoughts on the significance of these
biases and the challenges to mitigating or eliminating them.