method: Hi-VT5-beamsearch2023-04-18

Authors: JiangLong He, Mamatha N, Shiv Vignesh, Deepak Kumar

Description: Hi-VT5 model pretrained with private custom document collection using span masking objective. Pretrained model is then trained with DUDE dataset and Multi-Page DocVQA dataset.

Rubèn Tito, Dimosthenis Karatzas, Ernest Valveny: “Hierarchical multimodal transformers for Multi-Page DocVQA”, 2022; [http://arxiv.org/abs/2212.05935 arXiv:2212.05935].