microsoft / unilm

Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities
https://aka.ms/GeneralAI
MIT License
19.48k stars 2.48k forks source link

LayoutXML on DocVQA #618

Open oguz-akkas-deepsee opened 2 years ago

oguz-akkas-deepsee commented 2 years ago

Describe Model I am using (UniLM, MiniLM, LayoutLM ...): LayoutXML

I am trying to fine-tune LayoutXML model on DocVQA task. I am wondering if a document in a foreign language say in German, would my question and answer need to be in that language or it is okay to prepare data in English?

Thanks

wolfshow commented 2 years ago

@oguz-akkas-deepsee You may try both, but with sufficient German training samples, you will get better results.

oguz-akkas-deepsee commented 2 years ago

@oguz-akkas-deepsee You may try both, but with sufficient German training samples, you will get better results.

Thanks @wolfshow. I started to experiment that.