In this paper, we propose LayoutLMv3 to pre-train multimodal Transformers for Document AI with unified text and image masking. |
In this paper, we propose LayoutLMv3 to pre-train multimodal Transformers for Document AI with unified text and image masking. |