2020.3 Enhanced meta-learning for cross-lingual named entity recognition with minimal resources 閱讀筆記

Motivation Problem Setting: a) One source language with rich labeled data. b) No labeled data in the target language. 現有的 Cross-lingula NER 方法可以分爲兩大類: a) Label projection (generate labeled data in tar
相關文章
相關標籤/搜索