English

Improving Low-Resource Cross-lingual Parsing with Expected Statistic Regularization

Computation and Language 2022-10-19 v1

Abstract

We present Expected Statistic Regularization (ESR), a novel regularization technique that utilizes low-order multi-task structural statistics to shape model distributions for semi-supervised learning on low-resource datasets. We study ESR in the context of cross-lingual transfer for syntactic analysis (POS tagging and labeled dependency parsing) and present several classes of low-order statistic functions that bear on model behavior. Experimentally, we evaluate the proposed statistics with ESR for unsupervised transfer on 5 diverse target languages and show that all statistics, when estimated accurately, yield improvements to both POS and LAS, with the best statistic improving POS by +7.0 and LAS by +8.5 on average. We also present semi-supervised transfer and learning curve experiments that show ESR provides significant gains over strong cross-lingual-transfer-plus-fine-tuning baselines for modest amounts of label data. These results indicate that ESR is a promising and complementary approach to model-transfer approaches for cross-lingual parsing.

Keywords

Cite

@article{arxiv.2210.09428,
  title  = {Improving Low-Resource Cross-lingual Parsing with Expected Statistic Regularization},
  author = {Thomas Effland and Michael Collins},
  journal= {arXiv preprint arXiv:2210.09428},
  year   = {2022}
}

Comments

Accepted in TACL 2022, pre-MIT Press publication version

R2 v1 2026-06-28T03:51:56.251Z