Bias in Large Language Models
Talk, Dataminded Teamweekend, Bernkastel-Kaus, Germany
I was invited to give a keynote about social bias in large language models at the Dataminded team weekend.
Talks and presentations
Language Bias in AI: Ensuring Fairness Across Global English Speakers
Talk, Social Sciences and Humanities Research Day, Leuven, Belgium
I was invited to give a pitch about my research on language bias in AI at the Social Sciences and Humanities Research Day. The paper I talked about is available here.
LLMs for Keywording & Entity Recognition
Talk, AI in Cultural Heritage Group, online
I was invited to give a talk in the AI in Cultural Heritage Group on the recent publication of Large Language Models to make museum archive collections more accessible. The paper is available here
Native design bias in Large Language Models
Talk, ORBEL, Maastricht, Netherlands
I held a talk on native design bias in Large Language Models at ORBEL in Maastricht. The abstract is available here
GenAI: Introduction and Academic Perspective
Talk, Spring Fest, Leuven, Belgium
I held a talk on the academic perspective of GenAI at the alumni day for postgraduate studies in Big Data and Analytics at the KU Leuven. More specifically, on the transferability of debiasing in a multilingual language model. URL
Onderzoekssamenwerking tussen VSA en KU Leuven
Talk, Dag van het netwerk Statistiek Vlaanderen, Brussel, Belgium
I presented the results of the research collaboration between the VSA and the KU Leuven. This was a combination of work done by several thesis groups over the last years. The talk was held in Dutch. URL slides
Investigating Bias in Multilingual Language Models: Cross-Lingual Transfer of Debiasing Techniques
Conference proceedings talk, EMNLP, 8 Sentosa Gateway, Singapore
This was an (oral) presentation of the paper ‘Investigating Bias in Multilingual Language Models: Cross-Lingual Transfer of Debiasing Techniques’
Benchmarking methods on different text classification tasks.
Conference proceedings talk, IFORS, Santiago, Chile
I presented our paper ‘Evaluating text classification: A benchmark study’. Abstract
Benchmarking Dutch sentiment models for use in Twitter sentiment annotation
Talk, Statistics Flanders Seminar, Leuven, Belgium