1 paper
Hammad A. Ayyubi, Rahul Lokesh, Alireza Zareian +2
Image-caption pretraining has been quite successfully used for downstream vision tasks like zero-shot image classification and object detection. However, image-caption pretraining…