The IMS-CUBoulder System for the SIGMORPHON 2020 Shared Task on Unsupervised Morphological Paradigm Completion

TL;DR

This paper describes the IMS-CUBoulder system for unsupervised morphological paradigm completion, which improves over the baseline by using advanced neural models and achieves top performance in the SIGMORPHON 2020 shared task.

Contribution

The paper introduces a modified system with LSTM-based models that outperform the baseline in unsupervised morphological paradigm completion.

Findings

Pointer-generator model achieved the best overall score.

System outperformed the baseline on Bulgarian and Kannada.

Demonstrated effectiveness of neural sequence models in unsupervised morphology tasks.

Abstract

In this paper, we present the systems of the University of Stuttgart IMS and the University of Colorado Boulder (IMS-CUBoulder) for SIGMORPHON 2020 Task 2 on unsupervised morphological paradigm completion (Kann et al., 2020). The task consists of generating the morphological paradigms of a set of lemmas, given only the lemmas themselves and unlabeled text. Our proposed system is a modified version of the baseline introduced together with the task. In particular, we experiment with substituting the inflection generation component with an LSTM sequence-to-sequence model and an LSTM pointer-generator network. Our pointer-generator system obtains the best score of all seven submitted systems on average over all languages, and outperforms the official baseline, which was best overall, on Bulgarian and Kannada.