Hello,
Is there any explication why the results reported for Donut in the paper are much lower than those found in the original paper for DocVQA ?
Did you use a different resolution ? Or was the version used for Donut a previous one ?
Thanks !
The numbers in my paper were from the first version of the Donut paper put on arXiv. They released a newer version of the Donut paper later which had better results (from using real data in pre-training).
Hello, Is there any explication why the results reported for Donut in the paper are much lower than those found in the original paper for DocVQA ? Did you use a different resolution ? Or was the version used for Donut a previous one ? Thanks !