New Perspectives on Bias and Discrimination in Language Technology

4th & 5th of November, 2024 @University of Amsterdam

Workshop

One of the central issues discussed in the context of the societal impact of language technology is that machine learning systems can contribute to discrimination, for instance by propagating human biases and stereotypes. Despite efforts to address these issues, we are far from solving them.

The goal of this workshop is to bring together researchers from different fields to discuss the state of the art on bias measurement and mitigation in language technology and to explore new avenues of approach. For more information, read our Call for Abstracts.

The Call for Abstracts is currently closed!

Please check your spam folder if you haven't heard from us yet.

Organizers

This workshop is organized by Katrin Schulz, Leendert van Maanen, Jelle Zuidema, Oskar van der Wal and Dominik Bachmann as part of the project “The biased reality of online media - Using stereotypes to make media manipulation visible”, which is financed by the Dutch Research Council (NWO). In this project we integrate knowledge from AI, psychology and linguistics to develop measures for social biases in language models and humans, and then to use these to study the influence of our media consumption on our beliefs.

Important Dates

Deadline Call for Abstracts: 15 September, 2024
Notification of acceptance: 24 September, 2024
• Workshop: 4 & 5 November, 2024

Program (tentative)

This workshop will be held in Amsterdam. For questions or comments, please contact Oskar via o.d.vanderwal@uva.nl.

Monday, 4th of November (click to expand)
09:00 doors open, coffee
09:30 welcome
09:45 keynote 1: Dong Nguyen
11:00 coffee
11:15 Vera Neplenbroek, Arianna Bisazza and Raquel Fernández: MBBQ: A Dataset for Cross-Lingual Comparison of Stereotypes in Generative LLMs
11:45Beatrice Savoldi, Jasmijn Bastings, Luisa Bentivogli and Eva Vanmassenhove: A Decade of Gender Bias in Machine Translation
12:15lunch
13:45keynote 2: John Lalor
15:00coffee
15:30Flor Miriam Plaza del Arco, Amanda Cercas Curry, Alba Curry, Gavin Abercrombie and Dirk Hovy: Angry Men, Sad Women: Large Language Models Reflect Gendered Stereotypes in Emotion Attribution
16:00Hellina Hailu Nigatu and Zeerak Talat: A Capabilities Approach to Studying Bias and Harm in Language Technologies
16:30poster session with drinks
18:00end of day 1
Tuesday, 5th of November (click to expand)
09:00doors open, coffee
09:30keynote 3: Zeerak Talat
10:45coffee
11:00Paula Helm and Gabor Bella: Resisting language modeling bias through pluriversal language technology design
11:30Raquel Freitag and Tulio Gois: Sociolinguistic Discrimination in Large Language Models: A Comparative Analysis of Dialectal Variability in Brazilian Portuguese
12:00lunch
13:30panel discussion
14:15coffee
14:30keynote 4: Abigail Jacobs
15:45coffee
16:00seminar on AI and systemic injustice
18:00end of day 2
19:00dinner for everyone to join (but self-paid)

Invited Speakers

We are excited to confirm the following invited speakers at the workshop.

Dong Nguyen

Dr. Dong Nguyen is assistant professor Computer Science at Utrecht University. She works in the field of Natural Language Processing and is the head of the NLP and Society Lab. Her research focuses on computational text analysis for research questions from the social sciences. In recent years, she has also worked on bias measurement and transparent NLP. [Link to their website]

Monday 4th of November at 9:45.
John Lalor

Dr. John Lalor is assistant professor of IT, Analytics, and Operations at the University of Notre Dame, Indiana. His research focuses on developing methods for evaluating machine learning (and especially Natural Language Processing) models and for quantifying uncertainty. [Link to their website]

Monday 4th of November at 13:45.
Zeerak Talat

Dr. Zeerak Talat is a research fellow at Mohamed Bin Zayed University of AI, and in November, they will be a Chancellor's Fellow in Responsible ML and AI at the Centre for Technomoral ML and AI at the Edinburg Futures Institute and Institute for Language, Cognition, and Computation. Their research seeks to examine how machine learning systems interact with our societies and the downstream effects of introducing machine learning to our society. [Link to their website]

Tuesday 5th of November at 9:45.
Abigail Jacobs

Dr. Abigail Jacobs is assistant professor of Information at the University of Michigan in the School of Information and assistant professor of Complex Systems in the College of Literature, Science, and the Arts. Her current research interests are around measurement; the hidden assumptions in machine learning, focusing on measurement and validity as a lens; structure, governance, and inequality in sociotechnical systems; and social networks. [Link to their website]