[PDF]

Labeling and Quantifying Biases in Text Generated by Pre-trained Language Models


Xinru Wang

08/05/2025

Supervised by Nedjma Ousidhoum; Moderated by Jandson Santos Ribeiro Santos

In this work, the student will look into NLP tools and train models to study bias in text. The student will be initiated to different techniques commonly used in computational social science and natural language processing and will choose the kind of study they'd like to conduct in any language. They will then train a model for further analysis and share their insights.


Initial Plan (03/02/2025) [Zip Archive]

Final Report (08/05/2025) [Zip Archive]

Publication Form