Author & Affiliation: Wei Yi, principal data scientist at AstraZeneca with past experience in SecondMin, Microsoft Research and a hedge fund CTO role.
Content The article details an image-text multi-modality model that can handle various modalities such as text and images for tasks like classification, retrieval, and captioning.
Contributions to Field: Discusses the surge of multi-modality foundation models in data science applications requiring integrated knowledge from multiple datatypes.