Laure Delisle

Compositionality, Vision-Language Models, Representation Learning, Interpretability