English

Improving Self-training for Cross-lingual Named Entity Recognition with Contrastive and Prototype Learning

Computation and Language 2023-06-06 v2

Abstract

In cross-lingual named entity recognition (NER), self-training is commonly used to bridge the linguistic gap by training on pseudo-labeled target-language data. However, due to sub-optimal performance on target languages, the pseudo labels are often noisy and limit the overall performance. In this work, we aim to improve self-training for cross-lingual NER by combining representation learning and pseudo label refinement in one coherent framework. Our proposed method, namely ContProto mainly comprises two components: (1) contrastive self-training and (2) prototype-based pseudo-labeling. Our contrastive self-training facilitates span classification by separating clusters of different classes, and enhances cross-lingual transferability by producing closely-aligned representations between the source and target language. Meanwhile, prototype-based pseudo-labeling effectively improves the accuracy of pseudo labels during training. We evaluate ContProto on multiple transfer pairs, and experimental results show our method brings in substantial improvements over current state-of-the-art methods.

Keywords

Cite

@article{arxiv.2305.13628,
  title  = {Improving Self-training for Cross-lingual Named Entity Recognition with Contrastive and Prototype Learning},
  author = {Ran Zhou and Xin Li and Lidong Bing and Erik Cambria and Chunyan Miao},
  journal= {arXiv preprint arXiv:2305.13628},
  year   = {2023}
}

Comments

Accepted by ACL2023

R2 v1 2026-06-28T10:42:20.100Z