Authors: Baek Jeonghun, Baek Youngmin, Cha Junbum, Kim Seonghyeon, Lee Hwalsuk, Lee Sang-Woo, Lyu Sungwon, Yang Sohee, Yim Jinyeong

Description: MAC networks for combining visual cues and questions that embedded using BERTs, and pointer networks for pointing coordinates of text boxes that matches answers