Morphological Reinflection with Multiple Arguments: An Extended Annotation schema and a Georgian Case Study

TL;DR

This paper extends the UniMorph morphological annotation schema to better handle complex argument marking, demonstrates its application on Georgian, and shows that this improves dataset coverage and interpretability for morphological reinflection tasks.

Contribution

The paper introduces a hierarchical annotation schema for morphological data, applied to Georgian, significantly increasing dataset size and coverage, and analyzing the impact on reinflection model performance.

Findings

Extended schema captures complex argument marking effectively.

Georgian dataset size increased fourfold, verb forms sixfold.

Reinflection generalization is easier at form level than lemma level.

Abstract

In recent years, a flurry of morphological datasets had emerged, most notably UniMorph, a multi-lingual repository of inflection tables. However, the flat structure of the current morphological annotation schema makes the treatment of some languages quirky, if not impossible, specifically in cases of polypersonal agreement, where verbs agree with multiple arguments using true affixes. In this paper, we propose to address this phenomenon by expanding the UniMorph annotation schema to a hierarchical feature structure that naturally accommodates complex argument marking. We apply this extended schema to one such language, Georgian, and provide a human-verified, accurate and balanced morphological dataset for Georgian verbs. The dataset has 4 times more tables and 6 times more verb forms compared to the existing UniMorph dataset, covering all possible variants of argument marking, demonstrating the adequacy of our proposed scheme. Experiments with a standard reinflection model show that generalization is easy when the data is split at the form level, but extremely hard when splitting along lemma lines. Expanding the other languages in UniMorph to this schema is expected to improve both the coverage, consistency and interpretability of this benchmark.