Lung-Hao Lee, Man-Chen Hung, Chao-Yi Chen, Rou-An Chen, and Yuen-Hsien Tseng.
In Proceedings of the 29th International Conference on Computers in Education (ICCE’21), pages 111-113.
Abstract
We explore transformer-based neural networks for Chinese grammatical error detection. The TOCFL learner corpus is used to measure the model capability of indicating whether a sentence contains errors or not. Experimental results show that ELECTRA transformers which take into account both transformer architecture and adversarial learning technique can achieve promising effectiveness with an improvement of F1-score.
