Single-/Multi-Source Cross-Lingual NER via Teacher-Student Learning on Unlabeled Data in Target Language

To better tackle the named entity recognition (NER) problem on languages with little/no labeled data, cross-lingual NER must effectively leverage knowledge learned from source languages with rich labeled data. Previous works on cross-lingual NER are mostly based on label projection with pairwise texts or direct model transfer. However, such methods either are not applicable if the labeled data in the source languages is unavailable, or do not leverage information contained in unlabeled data in the target language. In this paper, we propose a teacher-student learning method to address such limitations, where NER models in the source languages are used as teachers to train a student model on unlabeled data in the target language. The proposed method works for both single-source and multi-source cross-lingual NER. For the latter, we further propose a similarity measuring method to better weight the supervision from different teacher models. Extensive experiments for 3 target languages on benchmark datasets well demonstrate that our method outperforms existing state-of-the-art methods for both single-source and multi-source cross-lingual NER.

PDF Abstract ACL 2020 PDF ACL 2020 Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Cross-Lingual NER CoNLL Dutch SMTS Multi sim F1 81.33 # 3
Cross-Lingual NER CoNLL Dutch SMTS Multi avg F1 80.7 # 5
Cross-Lingual NER CoNLL Dutch SMTS Single F1 80.89 # 4
Cross-Lingual NER CoNLL German SMTS Multi sim F1 75.33 # 1
Cross-Lingual NER CoNLL German SMTS Multi avg F1 74.97 # 2
Cross-Lingual NER CoNLL German SMTS Single F1 73.22 # 5
Cross-Lingual NER CoNLL Spanish SMTS Single F1 76.94 # 5
Cross-Lingual NER CoNLL Spanish SMTS Multi sim F1 78 # 3
Cross-Lingual NER CoNLL Spanish SMTS Multi avg F1 77.75 # 4

Methods


No methods listed for this paper. Add relevant methods here