Jointly Learning Span Extraction and Sequence Labeling for Information Extraction from Business Documents
This addresses the problem of extracting sparse information from long business documents for users in document processing, though it is incremental as it combines existing techniques.
The paper tackles information extraction from business documents by jointly learning span extraction and sequence labeling, resulting in promising performance and significantly faster speed than span-based methods on four datasets in English and Japanese.
This paper introduces a new information extraction model for business documents. Different from prior studies which only base on span extraction or sequence labeling, the model takes into account advantage of both span extraction and sequence labeling. The combination allows the model to deal with long documents with sparse information (the small amount of extracted information). The model is trained end-to-end to jointly optimize the two tasks in a unified manner. Experimental results on four business datasets in English and Japanese show that the model achieves promising results and is significantly faster than the normal span-based extraction method. The code is also available.