Project Description

In the past few years, the NLP group of the ERSP project has studied the broad area of NLP and ethics. In particular, we've studied gender bias and gender bias in information extraction (class of 2018-2019), and language / dialect / racial bias in modern pretrained language models (class of 2019-2020). The undergraduate students have published 2 papers at top venues with another 2 under review. In this new class of ERSP project, we will continue to work on issues in natural language processing related to ethical and social topics and the overall impact of these algorithms on people and society. Topics include: adversarial attacks, privacy and computational profiling, manipulation and framing, and other applications to social good.

Team Members

Professor and Mentors

Meeting Time

  • Meeting with Prof. Mirza and Prof. Eiers
    • Location: Zoom
    • Time: Wed, 3:30-4pm

Links to Proposals and Presentation

  • Proposal link
  • Final presentation:

Individual Logs

Peer Review

Project Documentation and Resources

Poster