Pushing the Limits of Low-Resource Morphological Inflection
This addresses the challenge of generating morphological inflections for languages with limited data, which is incremental as it builds on existing methods to enhance low-resource performance.
The paper tackles the problem of poor performance of neural methods in low-resource morphological inflection generation by proposing improvements including a novel two-step attention architecture and cross-lingual transfer strategies, resulting in a 15 percentage point increase in macro-averaged accuracy over the state-of-the-art.
Recent years have seen exceptional strides in the task of automatic morphological inflection generation. However, for a long tail of languages the necessary resources are hard to come by, and state-of-the-art neural methods that work well under higher resource settings perform poorly in the face of a paucity of data. In response, we propose a battery of improvements that greatly improve performance under such low-resource conditions. First, we present a novel two-step attention architecture for the inflection decoder. In addition, we investigate the effects of cross-lingual transfer from single and multiple languages, as well as monolingual data hallucination. The macro-averaged accuracy of our models outperforms the state-of-the-art by 15 percentage points. Also, we identify the crucial factors for success with cross-lingual transfer for morphological inflection: typological similarity and a common representation across languages.