Single-Model Encoder-Decoder with Explicit Morphological Representation for Reinflection
This work addresses the problem of morphological reinflection for low-resource languages, offering a more data-efficient solution.
The paper tackles morphological reinflection by proposing a neural encoder-decoder model with explicit morphological representation, which reduces training data needs and achieves state-of-the-art results, making it applicable to low-resource languages.
Morphological reinflection is the task of generating a target form given a source form, a source tag and a target tag. We propose a new way of modeling this task with neural encoder-decoder models. Our approach reduces the amount of required training data for this architecture and achieves state-of-the-art results, making encoder-decoder models applicable to morphological reinflection even for low-resource languages. We further present a new automatic correction method for the outputs based on edit trees.