Cross-lingual NER Transferability via Language Alignment Metrics

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 limi

Research goal: How does the alignment between source and target languages (e.g., via language similarity metrics like word embedding distances) affect the transferability of knowledge in teacher-student cross-lingual NER models, as measured by F1 scores on the CoNLL-2003 and WikiAnn benchmarks?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.8/10.