Understanding Transformers for Information Extraction with Limited Data

no code implementations • PACLIC 2020 • Minh-Tien Nguyen, Dung Tien Le, Nguyen Hong Son, Bui Cong Minh, Do Hoang Thai Duong, Le Thai Linh

Jointly Learning Span Extraction and Sequence Labeling for Information Extraction from Business Documents

no code implementations • 26 May 2022 • Nguyen Hong Son, Hieu M. Vu, Tuan-Anh D. Nguyen, Minh-Tien Nguyen

This paper introduces a new information extraction model for business documents.

Enhance Incomplete Utterance Restoration by Joint Learning Token Extraction and Text Generation

1 code implementation • NAACL 2022 • Shumpei Inoue, Tsungwei Liu, Nguyen Hong Son, Minh-Tien Nguyen

To support the picker, we design two label creation methods (soft and hard labels), which can work in cases of no annotation data for the omitted tokens.

Language Modelling Text Generation

A Span Extraction Approach for Information Extraction on Visually-Rich Documents

no code implementations • 2 Jun 2021 • Tuan-Anh D. Nguyen, Hieu M. Vu, Nguyen Hong Son, Minh-Tien Nguyen

Firstly, we introduce a new query-based IE model that employs span extraction instead of using the common sequence labeling approach.

Language Modelling

Transfer Learning for Information Extraction with Limited Data

no code implementations • 6 Mar 2020 • Minh-Tien Nguyen, Viet-Anh Phan, Le Thai Linh, Nguyen Hong Son, Le Tien Dung, Miku Hirano, Hajime Hotta

This paper presents a practical approach to fine-grained information extraction.

General Classification Transfer Learning