Masked Part-Of-Speech Model: Does Modeling Long Context Help Unsupervised POS-tagging?

TL;DR

This paper introduces MPoSM, a masked language model inspired approach for unsupervised POS tagging that models long-range dependencies, with mixed results across languages and insights from synthetic experiments.

Contribution

Proposes MPoSM, a novel masked language model-based framework for unsupervised POS induction capable of modeling arbitrary tag dependencies.

Findings

MPoSM achieves competitive results on multiple datasets.

Modeling long-term dependencies shows mixed effects across languages.

Synthetic experiments reveal challenges in learning simple tag agreements.

Abstract

Previous Part-Of-Speech (POS) induction models usually assume certain independence assumptions (e.g., Markov, unidirectional, local dependency) that do not hold in real languages. For example, the subject-verb agreement can be both long-term and bidirectional. To facilitate flexible dependency modeling, we propose a Masked Part-of-Speech Model (MPoSM), inspired by the recent success of Masked Language Models (MLM). MPoSM can model arbitrary tag dependency and perform POS induction through the objective of masked POS reconstruction. We achieve competitive results on both the English Penn WSJ dataset as well as the universal treebank containing 10 diverse languages. Though modeling the long-term dependency should ideally help this task, our ablation study shows mixed trends in different languages. To better understand this phenomenon, we design a novel synthetic experiment that can specifically diagnose the model's ability to learn tag agreement. Surprisingly, we find that even strong baselines fail to solve this problem consistently in a very simplified setting: the agreement between adjacent words. Nonetheless, MPoSM achieves overall better performance. Lastly, we conduct a detailed error analysis to shed light on other remaining challenges. Our code is available at https://github.com/owenzx/MPoSM