Pre-trained Model for Chinese Word Segmentation with Meta Learning

Recent researches show that pre-trained models such as BERT (Devlin et al., 2019) are beneficial for Chinese Word Segmentation tasks. However, existing approaches usually finetune pre-trained models directly on a separate downstream Chinese Word Segmentation corpus. These recent methods don't fully utilize the prior knowledge of existing segmentation corpora, and don't regard the discrepancy between the pre-training tasks and the downstream Chinese Word Segmentation tasks. In this work, we propose a Pre-Trained Model for Chinese Word Segmentation, which can be abbreviated as PTM-CWS. PTM-CWS model employs a unified architecture for different segmentation criteria, and is pre-trained on a joint multi-criteria corpus with meta learning algorithm. Empirical results show that our PTM-CWS model can utilize the existing prior segmentation knowledge, reduce the discrepancy between the pre-training tasks and the downstream Chinese Word Segmentation tasks, and achieve new state-of-the-art performance on twelve Chinese Word Segmentation corpora.

Unified Multi-Criteria Chinese Word Segmentation with BERT

Multi-Criteria Chinese Word Segmentation (MCCWS) aims at finding word boundaries in a Chinese sentence composed of continuous characters while multiple segmentation criteria exist. The unified framework has been widely used in MCCWS and shows its effectiveness. Besides, the pre-trained BERT language model has been also introduced into the MCCWS task in a multi-task learning framework. In this paper, we combine the superiority of the unified framework and pretrained language model, and propose a unified MCCWS model based on BERT. Moreover, we augment the unified BERT-based MCCWS model with the bigram features and an auxiliary criterion classification task. Experiments on eight datasets with diverse criteria demonstrate that our methods could achieve new state-of-the-art results for MCCWS.