Parcalabescu, Letitia (2024) Measuring the Contributions of Vision and Text Modalities in Multimodal Transformers. [Dissertation]