<ul><li>AxBERT is an interpretable deep learning model proposed for Chinese spelling correction.</li><li>It aligns with an associative knowledge network (AKN) constructed based on co-occurrence relations among Chinese characters.</li><li>A translator matrix between BERT and AKN is introduced for alignment and regulation of the attention component in BERT.</li><li>Experimental results on SIGHAN datasets show that AxBERT achieves extraordinary performance and interpretability.</li></ul>

AxBERT: An Interpretable Chinese Spelling Correction Method Driven by Associative Knowledge Network

Discover more