Robust Multilingual Part-of-Speech Tagging via Adversarial Training

2017/11/14 by Michihiro Yasunaga, Jungo Kasai, Yasunaga, Michihiro +3 · 1 citation
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Multimodal Machine Learning Applications #Natural Language Processing Techniques #Topic Modeling

paper · pdf · doi:10.48550/arxiv.1711.04903

openalex publication_date 2017/11/14 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Adversarial training (AT) is a powerful regularization method for neural networks, aiming to achieve robustness to input perturbations. Yet, the specific effects of the robustness obtained from AT are still unclear in the context of natural language processing. In this paper, we propose and analyze a neural POS tagging model that exploits AT. In our experiments on the Penn Treebank WSJ corpus and the Universal Dependencies (UD) dataset (27 languages), we find that AT not only improves the overall tagging accuracy, but also 1) prevents over-fitting well in low resource languages and 2) boosts tagging accuracy for rare / unseen words. We also demonstrate that 3) the improved tagging performance by AT contributes to the downstream task of dependency parsing, and that 4) AT helps the model to learn cleaner word representations. 5) The proposed AT model is generally effective in different sequence labeling tasks. These positive results motivate further use of AT for natural language tasks.

Citations

• Long Short-Term Memory
• Bidirectional recurrent neural networks
• Data Noising as Smoothing in Neural Network Language Models
• Adversarial Examples for Evaluating Reading Comprehension Systems
• Finding Function in Form: Compositional Character Models for Open Vocabulary Word Representation
• Transfer Learning for Sequence Tagging with Hierarchical Recurrent Networks

Cited by

• Fake News Classification in Urdu: A Domain Adaptation Approach for a Low-Resource Language

Related

• Adversarial Neural Networks for Cross-lingual Sequence Tagging
• The Best of Both Worlds: Lexical Resources To Improve Low-Resource Part-of-Speech Tagging
• Is POS Tagging Necessary or Even Helpful for Neural Dependency Parsing?
• Domain adaptation for part-of-speech tagging of noisy user-generated\n text
• Distant Supervision from Disparate Sources for Low-Resource\n Part-of-Speech Tagging