LayoutLMV3 on dataset other than english

I have a Japanese annotated receipt dataset that I want to train on LayoutLMv3 to detect and classify fields on the receipt. Does LayoutLMv3 accept non-English datasets?