|Sep. '22||New paper on arXiv! Causal Proxy Models for Concept-Based Model Explanations|
|Sep. '22||Paper (CEBaB) got accepted to NeurIPS 2022!|
|July '22||Gave a talk about CEBaB at the Stanford NLP Group. Slides available here. Video available here.|
|May '22||New paper on arXiv! CEBaB: Estimating the Causal Effects of Real-World Concepts on NLP Model Behavior.|
|Jan. '22||I'm currently at the Stanford NLP group for one year under guidance of prof. Christopher Potts.|
|Nov. '21||Started my Ph.D. at Ghent University.|
Hi! I’m a first-year PhD student passionate about Machine Learning, Natural Language Processing and explainable/causal AI. I work at the Text to Knowledge research group at Ghent University under supervision of prof. Chris Develder and prof. Thomas Demeester. I'm currently at the Stanford NLP group for one year under guidance of prof. Christopher Potts.
Currently, my research focusses on explainable, causally-motivated AI. Applying ideas from causal inference to machine learning, I aim to develop more interpretable and robust models. My overall goal is to build techniques that allow for intuitive inspection and editing of model behavior. At the Stanford NLP group, I've been working on a causally-motivated benchmark for interpretability techniques. Currently, we are using this new resource to guide the development of new state-of-the-art model explanation techniques. View my initial work.
Some other hobbies include cooking, camping and occasionally skiing. Some recent books I enjoyed include The Worldly Philosophers, Factfulness, 21 lessons for the 21st century, Sapiens, and Educated Previously, I spent a lot of my time practicing music. I play classical guitar and I sung in a choir, which gave me the opportunity to visit Mexico, Japan, Bulgaria, and Rome.