English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Pretrained language models (PLMs) have shown marvelous improvements acrossvarious NLP tasks. Most Chinese PLMs simply treat an input text as a sequenceof characters, and completely ignore word information. Although Whole WordMasking can alleviate this, the semantics in words is still not wellrepresented. In this paper, we revisit the segmentation granularity of ChinesePLMs. We propose a mixed-granularity Chinese BERT (MigBERT) by considering bothcharacters and words. To achieve this, we design objective functions forlearning both character and word-level representations. We conduct extensiveexperiments on various Chinese NLP tasks to evaluate existing PLMs as well asthe proposed MigBERT. Experimental results show that MigBERT achieves new SOTAperformance on all these tasks. Further analysis demonstrates that words aresemantically richer than characters. More interestingly, we show that MigBERTalso works with Japanese. Our code and model have been releasedhere~\footnote{https://github.com/xnliang98/MigBERT}.

Character, Word, or Both? Revisiting the Segmentation Granularity for Chinese Pre-trained Language Models