Authors: Tao Huang, Jie Wang, Tao Xu
Affiliation: Ant Group
Description: End-to-End OCR free based transformer for document understanding. The encoder maps a document image into embeddings and the decoder generates a sequence of tokens by the encoded embeddings, where the tokens can be converted into a kv type of entity linking in a structured form string. All the trained data is used and epoch is 900, and multiline keys are splited. The final result is corrected by the ocr output.