Keynotes

Sabine Weber: Queer NLP - Past, Present and Future

Bio: Dr. Sabine Weber is a postdoctoral researcher at the Bamberg NLP Group, specializing in domain adaptation, multilinguality and subjective tasks. They earned their Ph.D. in Computer Science from the University on Edinburgh, focusing on multilingual entailment detection. Before that they received a B.Sc. and a M.Sc. in Computer Engineering from TU Berlin, and a B.A. in Slavic Languages from the Humboldt University Berlin. In addition to building models that are representative of human diversity, they are interested in examining the societal impact of NLP. They are also a core organizer for the volunteer-run organization Queer in AI.

Abstract: Natural language processing (NLP) technologies are rapidly reshaping how language is created, processed, and interpreted by humans, touching nearly every aspect of life. In this talk I will give an overview of the relationship between LGBTQIA+ communities and NLP technologies, building upon a survey of all papers on the topic published in the ACL anthology (the main publication outlet for NLP research). This talk will shine a light on common topics, popular methods and existing gaps, and ask the question: Where should we go next?

Morgan Klaus Scheuerman - Envisioning Gender: The Social Production of Identity in Computer Vision

Bio: Morgan Klaus Scheuerman is a Research Scientist specializing in Human-Computer Interaction at Sony. He received his PhD in Information Science at University of Colorado Boulder, where he was a 2021 MSR Research Fellow. His research focuses on the intersection of technical infrastructure and marginalized identities. In particular, he examines how gender and race characteristics are embedded into algorithmic infrastructures and how those permeations influence the entire system. His work has received multiple best paper awards and honorable mentions at CHI and CSCW. He earned his MS degree in Human-Centered Computing from University of Maryland Baltimore County and his BA in Communication & Media Studies (Minor Gender & Sexuality Studies) from Goucher College.

Abstract: Computer vision technologies have been increasingly scrutinized in recent years for their propensity to cause harm. However, human identity is embedded into computer vision not only across these artifacts, but also across the network of human workers who shape computer vision systems. Given computer vision is designed by humans, there is ample opportunity to understand how human positionality influences the outcomes of computer vision systems. In this talk, I present the span of work I have conducted on how gender is implemented in computer vision: from how trans individuals perceive automated gender classification to how gender is represented in models and datasets to how different worker positionalities influence the development process. I showcase how representations of gender in computer vision have been and continue to be exclusionary and problematic. I highlight how traditional tech workers enact a positional power over data workers in the global south to continue these identity classification practices. I demonstrate how approaches to gender identity in computer vision thus move from something more open, contextual, and exploratory to a completely closed, binary, and prescriptive classification. I conclude with an agenda for expanding computational approaches to identity back towards something open, flexible, and situated.

How to reach Cartesium and SFG?

To reach the Cartesium and SFG building (Enrique-Schmidt-Straße 5) at the University of Bremen, please board Tram Line 6 (direction "Universität") from Bremen Hauptbahnhof or Bremen Airport. Exit the tram at the "Universität-Süd" stop. The Cartesium is the prominent red building located immediately east of the platform and SFG is a curved building beside it.

Organizers

Andrea Sipos She is a co-founder of Queer in Robotics. She leads a tactile sening group within the Perception and Cognition Department (PEK) at the Robotics and Mechatronics Center of the German Aerospace Center (DLR) in Munich, Germany. 

Pranav A He is a co-founder of Queer in AI, currently a PhD student at University of Hamburg, researching on AI policymaking.

Martin Mundt He is the professor for lifelong machine learning at the University of Bremen, leading the OWL-ML lab.